storage redundancy in Ovirt
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them. The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage? Only glusterfs? Thanks
You can use FC storage where luns come from a redundant storage array. On Apr 15, 2017 11:25, "Konstantin Raskoshnyi" <konrasko@gmail.com> wrote: Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them. The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage? Only glusterfs? Thanks _______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type. Fernando 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
In ovirt you have to attach storage through specific host. If host goes down storage is not available. On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Hello Konstantin. That doesn`t make much sense make a whole cluster depend on a single host.
From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
I think he means SPM? I've seen when node with SPM goes down it isn't really seamless and takes a minute or two to catch up unless of course you out that node in maintenance mode but that isn't possible if it crashes or something On Apr 15, 2017 1:38 PM, "FERNANDO FREDIANI" <fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage. On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted. Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com
wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote: On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted. In a properly configured setup the SBA should take care of that. That's the whole point of HA services Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com
wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically. Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
> Hi guys, > I have one nfs storage, > it's connected through host1. > host2 also has access to it, I can easily migrate vms between them. > > The question is - if host1 is down - all infrastructure is down, > since all traffic goes through host1, > is there any way in oVirt to use redundant storage? > > Only glusterfs? > > Thanks > > > _______________________________________________ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > >
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
So what's the whole HA point then? On Sun, Apr 16, 2017 at 4:29 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
> Well, make it not go through host1 and dedicate a storage server for > running NFS and make both hosts connect to it. > In my view NFS is much easier to manage than any other type of > storage, specially FC and iSCSI and performance is pretty much the same, so > you won`t get better results other than management going to other type. > > Fernando > > 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com> > : > >> Hi guys, >> I have one nfs storage, >> it's connected through host1. >> host2 also has access to it, I can easily migrate vms between them. >> >> The question is - if host1 is down - all infrastructure is down, >> since all traffic goes through host1, >> is there any way in oVirt to use redundant storage? >> >> Only glusterfs? >> >> Thanks >> >> >> _______________________________________________ >> Users mailing list >> Users@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> >>
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
> Well, make it not go through host1 and dedicate a storage server for > running NFS and make both hosts connect to it. > In my view NFS is much easier to manage than any other type of > storage, specially FC and iSCSI and performance is pretty much the same, so > you won`t get better results other than management going to other type. > > Fernando > > 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com> > : > >> Hi guys, >> I have one nfs storage, >> it's connected through host1. >> host2 also has access to it, I can easily migrate vms between them. >> >> The question is - if host1 is down - all infrastructure is down, >> since all traffic goes through host1, >> is there any way in oVirt to use redundant storage? >> >> Only glusterfs? >> >> Thanks >> >> >> _______________________________________________ >> Users mailing list >> Users@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> >>
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node. On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com> :
> In ovirt you have to attach storage through specific host. > If host goes down storage is not available. > > On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < > fernando.frediani@upx.com> wrote: > >> Well, make it not go through host1 and dedicate a storage server >> for running NFS and make both hosts connect to it. >> In my view NFS is much easier to manage than any other type of >> storage, specially FC and iSCSI and performance is pretty much the same, so >> you won`t get better results other than management going to other type. >> >> Fernando >> >> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com >> >: >> >>> Hi guys, >>> I have one nfs storage, >>> it's connected through host1. >>> host2 also has access to it, I can easily migrate vms between them. >>> >>> The question is - if host1 is down - all infrastructure is down, >>> since all traffic goes through host1, >>> is there any way in oVirt to use redundant storage? >>> >>> Only glusterfs? >>> >>> Thanks >>> >>> >>> _______________________________________________ >>> Users mailing list >>> Users@ovirt.org >>> http://lists.ovirt.org/mailman/listinfo/users >>> >>>
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
> Hello Konstantin. > > That doesn`t make much sense make a whole cluster depend on a single > host. From what I know any host talk directly to NFS Storage Array or > whatever other Shared Storage you have. > Have you tested that host going down if that affects the other with > the NFS mounted directlly in a NFS Storage array ? > > Fernando > > 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com > >: > >> In ovirt you have to attach storage through specific host. >> If host goes down storage is not available. >> >> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >> fernando.frediani@upx.com> wrote: >> >>> Well, make it not go through host1 and dedicate a storage server >>> for running NFS and make both hosts connect to it. >>> In my view NFS is much easier to manage than any other type of >>> storage, specially FC and iSCSI and performance is pretty much the same, so >>> you won`t get better results other than management going to other type. >>> >>> Fernando >>> >>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>> konrasko@gmail.com>: >>> >>>> Hi guys, >>>> I have one nfs storage, >>>> it's connected through host1. >>>> host2 also has access to it, I can easily migrate vms between >>>> them. >>>> >>>> The question is - if host1 is down - all infrastructure is down, >>>> since all traffic goes through host1, >>>> is there any way in oVirt to use redundant storage? >>>> >>>> Only glusterfs? >>>> >>>> Thanks >>>> >>>> >>>> _______________________________________________ >>>> Users mailing list >>>> Users@ovirt.org >>>> http://lists.ovirt.org/mailman/listinfo/users >>>> >>>> > _______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Fence agent under each node? On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
> Hi Fernando, > I see each host has direct connection nfs mount, but yes, if main > host to which I connected nfs storage going down the storage becomes > unavailable and all vms are down > > > On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < > fernando.frediani@upx.com> wrote: > >> Hello Konstantin. >> >> That doesn`t make much sense make a whole cluster depend on a >> single host. From what I know any host talk directly to NFS Storage Array >> or whatever other Shared Storage you have. >> Have you tested that host going down if that affects the other with >> the NFS mounted directlly in a NFS Storage array ? >> >> Fernando >> >> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >> konrasko@gmail.com>: >> >>> In ovirt you have to attach storage through specific host. >>> If host goes down storage is not available. >>> >>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>> fernando.frediani@upx.com> wrote: >>> >>>> Well, make it not go through host1 and dedicate a storage server >>>> for running NFS and make both hosts connect to it. >>>> In my view NFS is much easier to manage than any other type of >>>> storage, specially FC and iSCSI and performance is pretty much the same, so >>>> you won`t get better results other than management going to other type. >>>> >>>> Fernando >>>> >>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>> konrasko@gmail.com>: >>>> >>>>> Hi guys, >>>>> I have one nfs storage, >>>>> it's connected through host1. >>>>> host2 also has access to it, I can easily migrate vms between >>>>> them. >>>>> >>>>> The question is - if host1 is down - all infrastructure is down, >>>>> since all traffic goes through host1, >>>>> is there any way in oVirt to use redundant storage? >>>>> >>>>> Only glusterfs? >>>>> >>>>> Thanks >>>>> >>>>> >>>>> _______________________________________________ >>>>> Users mailing list >>>>> Users@ovirt.org >>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>> >>>>> >> > _______________________________________________ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > > _______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
> When you set up a storage domain, you need to specify a host to > perform the initial storage operations, but once the SD is defined, it's > details are in the engine database, and all the hosts get connected to it > directly. If the first host you used to define the SD goes down, all other > hosts will still remain connected and work. SPM is an HA service, and if > the current SPM host goes down, SPM gets started on another host in the DC. > In short, unless your actual NFS exporting host goes down, there is no > outage. >
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
> > On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < > konrasko@gmail.com> wrote: > >> Hi Fernando, >> I see each host has direct connection nfs mount, but yes, if main >> host to which I connected nfs storage going down the storage becomes >> unavailable and all vms are down >> >> >> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >> fernando.frediani@upx.com> wrote: >> >>> Hello Konstantin. >>> >>> That doesn`t make much sense make a whole cluster depend on a >>> single host. From what I know any host talk directly to NFS Storage Array >>> or whatever other Shared Storage you have. >>> Have you tested that host going down if that affects the other >>> with the NFS mounted directlly in a NFS Storage array ? >>> >>> Fernando >>> >>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>> konrasko@gmail.com>: >>> >>>> In ovirt you have to attach storage through specific host. >>>> If host goes down storage is not available. >>>> >>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>> fernando.frediani@upx.com> wrote: >>>> >>>>> Well, make it not go through host1 and dedicate a storage server >>>>> for running NFS and make both hosts connect to it. >>>>> In my view NFS is much easier to manage than any other type of >>>>> storage, specially FC and iSCSI and performance is pretty much the same, so >>>>> you won`t get better results other than management going to other type. >>>>> >>>>> Fernando >>>>> >>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>> konrasko@gmail.com>: >>>>> >>>>>> Hi guys, >>>>>> I have one nfs storage, >>>>>> it's connected through host1. >>>>>> host2 also has access to it, I can easily migrate vms between >>>>>> them. >>>>>> >>>>>> The question is - if host1 is down - all infrastructure is >>>>>> down, since all traffic goes through host1, >>>>>> is there any way in oVirt to use redundant storage? >>>>>> >>>>>> Only glusterfs? >>>>>> >>>>>> Thanks >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Users mailing list >>>>>> Users@ovirt.org >>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>> >>>>>> >>> >> _______________________________________________ >> Users mailing list >> Users@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> >> > _______________________________________________ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users >
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks! On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
> > > On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote: > > On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote: > >> When you set up a storage domain, you need to specify a host to >> perform the initial storage operations, but once the SD is defined, it's >> details are in the engine database, and all the hosts get connected to it >> directly. If the first host you used to define the SD goes down, all other >> hosts will still remain connected and work. SPM is an HA service, and if >> the current SPM host goes down, SPM gets started on another host in the DC. >> In short, unless your actual NFS exporting host goes down, there is no >> outage. >> > > There is no storage outage, but if you shutdown the spm host, the > spm host > will not move to a new host until the spm host is online again, or > you confirm > manually that the spm host was rebooted. > > > In a properly configured setup the SBA should take care of that. > That's the whole point of HA services >
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
> > > Nir > > >> >> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >> konrasko@gmail.com> wrote: >> >>> Hi Fernando, >>> I see each host has direct connection nfs mount, but yes, if main >>> host to which I connected nfs storage going down the storage becomes >>> unavailable and all vms are down >>> >>> >>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>> fernando.frediani@upx.com> wrote: >>> >>>> Hello Konstantin. >>>> >>>> That doesn`t make much sense make a whole cluster depend on a >>>> single host. From what I know any host talk directly to NFS Storage Array >>>> or whatever other Shared Storage you have. >>>> Have you tested that host going down if that affects the other >>>> with the NFS mounted directlly in a NFS Storage array ? >>>> >>>> Fernando >>>> >>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>> konrasko@gmail.com>: >>>> >>>>> In ovirt you have to attach storage through specific host. >>>>> If host goes down storage is not available. >>>>> >>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>> fernando.frediani@upx.com> wrote: >>>>> >>>>>> Well, make it not go through host1 and dedicate a storage >>>>>> server for running NFS and make both hosts connect to it. >>>>>> In my view NFS is much easier to manage than any other type of >>>>>> storage, specially FC and iSCSI and performance is pretty much the same, so >>>>>> you won`t get better results other than management going to other type. >>>>>> >>>>>> Fernando >>>>>> >>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>> konrasko@gmail.com>: >>>>>> >>>>>>> Hi guys, >>>>>>> I have one nfs storage, >>>>>>> it's connected through host1. >>>>>>> host2 also has access to it, I can easily migrate vms between >>>>>>> them. >>>>>>> >>>>>>> The question is - if host1 is down - all infrastructure is >>>>>>> down, since all traffic goes through host1, >>>>>>> is there any way in oVirt to use redundant storage? >>>>>>> >>>>>>> Only glusterfs? >>>>>>> >>>>>>> Thanks >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> Users mailing list >>>>>>> Users@ovirt.org >>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>> >>>>>>> >>>> >>> _______________________________________________ >>> Users mailing list >>> Users@ovirt.org >>> http://lists.ovirt.org/mailman/listinfo/users >>> >>> >> _______________________________________________ >> Users mailing list >> Users@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> > > _______________________________________________ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > > >
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
> On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote: > >> >> >> On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote: >> >> On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote: >> >>> When you set up a storage domain, you need to specify a host to >>> perform the initial storage operations, but once the SD is defined, it's >>> details are in the engine database, and all the hosts get connected to it >>> directly. If the first host you used to define the SD goes down, all other >>> hosts will still remain connected and work. SPM is an HA service, and if >>> the current SPM host goes down, SPM gets started on another host in the DC. >>> In short, unless your actual NFS exporting host goes down, there is no >>> outage. >>> >> >> There is no storage outage, but if you shutdown the spm host, the >> spm host >> will not move to a new host until the spm host is online again, or >> you confirm >> manually that the spm host was rebooted. >> >> >> In a properly configured setup the SBA should take care of that. >> That's the whole point of HA services >> > > In some cases like power loss or hardware failure, there is no way > to start > the spm host, and the system cannot recover automatically. >
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
> > Nir > > >> >> >> Nir >> >> >>> >>> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >>> konrasko@gmail.com> wrote: >>> >>>> Hi Fernando, >>>> I see each host has direct connection nfs mount, but yes, if main >>>> host to which I connected nfs storage going down the storage becomes >>>> unavailable and all vms are down >>>> >>>> >>>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>>> fernando.frediani@upx.com> wrote: >>>> >>>>> Hello Konstantin. >>>>> >>>>> That doesn`t make much sense make a whole cluster depend on a >>>>> single host. From what I know any host talk directly to NFS Storage Array >>>>> or whatever other Shared Storage you have. >>>>> Have you tested that host going down if that affects the other >>>>> with the NFS mounted directlly in a NFS Storage array ? >>>>> >>>>> Fernando >>>>> >>>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>>> konrasko@gmail.com>: >>>>> >>>>>> In ovirt you have to attach storage through specific host. >>>>>> If host goes down storage is not available. >>>>>> >>>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>>> fernando.frediani@upx.com> wrote: >>>>>> >>>>>>> Well, make it not go through host1 and dedicate a storage >>>>>>> server for running NFS and make both hosts connect to it. >>>>>>> In my view NFS is much easier to manage than any other type of >>>>>>> storage, specially FC and iSCSI and performance is pretty much the same, so >>>>>>> you won`t get better results other than management going to other type. >>>>>>> >>>>>>> Fernando >>>>>>> >>>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>>> konrasko@gmail.com>: >>>>>>> >>>>>>>> Hi guys, >>>>>>>> I have one nfs storage, >>>>>>>> it's connected through host1. >>>>>>>> host2 also has access to it, I can easily migrate vms between >>>>>>>> them. >>>>>>>> >>>>>>>> The question is - if host1 is down - all infrastructure is >>>>>>>> down, since all traffic goes through host1, >>>>>>>> is there any way in oVirt to use redundant storage? >>>>>>>> >>>>>>>> Only glusterfs? >>>>>>>> >>>>>>>> Thanks >>>>>>>> >>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Users mailing list >>>>>>>> Users@ovirt.org >>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>> >>>>>>>> >>>>> >>>> _______________________________________________ >>>> Users mailing list >>>> Users@ovirt.org >>>> http://lists.ovirt.org/mailman/listinfo/users >>>> >>>> >>> _______________________________________________ >>> Users mailing list >>> Users@ovirt.org >>> http://lists.ovirt.org/mailman/listinfo/users >>> >> >> _______________________________________________ >> Users mailing list >> Users@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> >> >>
Oh, fence agent works fine if I select ilo4, Thank you for your help! On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
> On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> > wrote: > >> On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> >> wrote: >> >>> >>> >>> On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote: >>> >>> On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> >>> wrote: >>> >>>> When you set up a storage domain, you need to specify a host to >>>> perform the initial storage operations, but once the SD is defined, it's >>>> details are in the engine database, and all the hosts get connected to it >>>> directly. If the first host you used to define the SD goes down, all other >>>> hosts will still remain connected and work. SPM is an HA service, and if >>>> the current SPM host goes down, SPM gets started on another host in the DC. >>>> In short, unless your actual NFS exporting host goes down, there is no >>>> outage. >>>> >>> >>> There is no storage outage, but if you shutdown the spm host, the >>> spm host >>> will not move to a new host until the spm host is online again, or >>> you confirm >>> manually that the spm host was rebooted. >>> >>> >>> In a properly configured setup the SBA should take care of that. >>> That's the whole point of HA services >>> >> >> In some cases like power loss or hardware failure, there is no way >> to start >> the spm host, and the system cannot recover automatically. >> > > There are always corner cases, no doubt. But in a normal situation. > where an SPM host goes down because of a hardware failure, it gets fenced, > other hosts contend for SPM and start it. No surprises there. > > >> >> Nir >> >> >>> >>> >>> Nir >>> >>> >>>> >>>> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >>>> konrasko@gmail.com> wrote: >>>> >>>>> Hi Fernando, >>>>> I see each host has direct connection nfs mount, but yes, if >>>>> main host to which I connected nfs storage going down the storage becomes >>>>> unavailable and all vms are down >>>>> >>>>> >>>>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>>>> fernando.frediani@upx.com> wrote: >>>>> >>>>>> Hello Konstantin. >>>>>> >>>>>> That doesn`t make much sense make a whole cluster depend on a >>>>>> single host. From what I know any host talk directly to NFS Storage Array >>>>>> or whatever other Shared Storage you have. >>>>>> Have you tested that host going down if that affects the other >>>>>> with the NFS mounted directlly in a NFS Storage array ? >>>>>> >>>>>> Fernando >>>>>> >>>>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>>>> konrasko@gmail.com>: >>>>>> >>>>>>> In ovirt you have to attach storage through specific host. >>>>>>> If host goes down storage is not available. >>>>>>> >>>>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>>>> fernando.frediani@upx.com> wrote: >>>>>>> >>>>>>>> Well, make it not go through host1 and dedicate a storage >>>>>>>> server for running NFS and make both hosts connect to it. >>>>>>>> In my view NFS is much easier to manage than any other type >>>>>>>> of storage, specially FC and iSCSI and performance is pretty much the same, >>>>>>>> so you won`t get better results other than management going to other type. >>>>>>>> >>>>>>>> Fernando >>>>>>>> >>>>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>> konrasko@gmail.com>: >>>>>>>> >>>>>>>>> Hi guys, >>>>>>>>> I have one nfs storage, >>>>>>>>> it's connected through host1. >>>>>>>>> host2 also has access to it, I can easily migrate >>>>>>>>> vms between them. >>>>>>>>> >>>>>>>>> The question is - if host1 is down - all infrastructure is >>>>>>>>> down, since all traffic goes through host1, >>>>>>>>> is there any way in oVirt to use redundant storage? >>>>>>>>> >>>>>>>>> Only glusterfs? >>>>>>>>> >>>>>>>>> Thanks >>>>>>>>> >>>>>>>>> >>>>>>>>> _______________________________________________ >>>>>>>>> Users mailing list >>>>>>>>> Users@ovirt.org >>>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>>> >>>>>>>>> >>>>>> >>>>> _______________________________________________ >>>>> Users mailing list >>>>> Users@ovirt.org >>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>> >>>>> >>>> _______________________________________________ >>>> Users mailing list >>>> Users@ovirt.org >>>> http://lists.ovirt.org/mailman/listinfo/users >>>> >>> >>> _______________________________________________ >>> Users mailing list >>> Users@ovirt.org >>> http://lists.ovirt.org/mailman/listinfo/users >>> >>> >>>
But actually, it didn't work well. After main SPM host went down I see this 2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection. So that means only if BMC is up it's possible to automatically switch SPM host? Thanks On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4, Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
> "Corner cases"? > I tried to simulate crash of SPM server and ovirt kept trying to > reistablished connection to the failed node. >
Did you configure fencing?
> > > On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote: > >> On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> >> wrote: >> >>> On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> >>> wrote: >>> >>>> >>>> >>>> On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote: >>>> >>>> On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> >>>> wrote: >>>> >>>>> When you set up a storage domain, you need to specify a host to >>>>> perform the initial storage operations, but once the SD is defined, it's >>>>> details are in the engine database, and all the hosts get connected to it >>>>> directly. If the first host you used to define the SD goes down, all other >>>>> hosts will still remain connected and work. SPM is an HA service, and if >>>>> the current SPM host goes down, SPM gets started on another host in the DC. >>>>> In short, unless your actual NFS exporting host goes down, there is no >>>>> outage. >>>>> >>>> >>>> There is no storage outage, but if you shutdown the spm host, the >>>> spm host >>>> will not move to a new host until the spm host is online again, >>>> or you confirm >>>> manually that the spm host was rebooted. >>>> >>>> >>>> In a properly configured setup the SBA should take care of that. >>>> That's the whole point of HA services >>>> >>> >>> In some cases like power loss or hardware failure, there is no way >>> to start >>> the spm host, and the system cannot recover automatically. >>> >> >> There are always corner cases, no doubt. But in a normal situation. >> where an SPM host goes down because of a hardware failure, it gets fenced, >> other hosts contend for SPM and start it. No surprises there. >> >> >>> >>> Nir >>> >>> >>>> >>>> >>>> Nir >>>> >>>> >>>>> >>>>> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >>>>> konrasko@gmail.com> wrote: >>>>> >>>>>> Hi Fernando, >>>>>> I see each host has direct connection nfs mount, but yes, if >>>>>> main host to which I connected nfs storage going down the storage becomes >>>>>> unavailable and all vms are down >>>>>> >>>>>> >>>>>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>>>>> fernando.frediani@upx.com> wrote: >>>>>> >>>>>>> Hello Konstantin. >>>>>>> >>>>>>> That doesn`t make much sense make a whole cluster depend on a >>>>>>> single host. From what I know any host talk directly to NFS Storage Array >>>>>>> or whatever other Shared Storage you have. >>>>>>> Have you tested that host going down if that affects the other >>>>>>> with the NFS mounted directlly in a NFS Storage array ? >>>>>>> >>>>>>> Fernando >>>>>>> >>>>>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>>>>> konrasko@gmail.com>: >>>>>>> >>>>>>>> In ovirt you have to attach storage through specific host. >>>>>>>> If host goes down storage is not available. >>>>>>>> >>>>>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>>>>> fernando.frediani@upx.com> wrote: >>>>>>>> >>>>>>>>> Well, make it not go through host1 and dedicate a storage >>>>>>>>> server for running NFS and make both hosts connect to it. >>>>>>>>> In my view NFS is much easier to manage than any other type >>>>>>>>> of storage, specially FC and iSCSI and performance is pretty much the same, >>>>>>>>> so you won`t get better results other than management going to other type. >>>>>>>>> >>>>>>>>> Fernando >>>>>>>>> >>>>>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>>> konrasko@gmail.com>: >>>>>>>>> >>>>>>>>>> Hi guys, >>>>>>>>>> I have one nfs storage, >>>>>>>>>> it's connected through host1. >>>>>>>>>> host2 also has access to it, I can easily migrate >>>>>>>>>> vms between them. >>>>>>>>>> >>>>>>>>>> The question is - if host1 is down - all infrastructure is >>>>>>>>>> down, since all traffic goes through host1, >>>>>>>>>> is there any way in oVirt to use redundant storage? >>>>>>>>>> >>>>>>>>>> Only glusterfs? >>>>>>>>>> >>>>>>>>>> Thanks >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> _______________________________________________ >>>>>>>>>> Users mailing list >>>>>>>>>> Users@ovirt.org >>>>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>>>> >>>>>>>>>> >>>>>>> >>>>>> _______________________________________________ >>>>>> Users mailing list >>>>>> Users@ovirt.org >>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> Users mailing list >>>>> Users@ovirt.org >>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>> >>>> >>>> _______________________________________________ >>>> Users mailing list >>>> Users@ovirt.org >>>> http://lists.ovirt.org/mailman/listinfo/users >>>> >>>> >>>>
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this [image: Screen Shot 2017-04-16 at 10.22.00 PM.png]
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC? If your SPM is no responsive, the system will try to fence it. Did you configure power management for all hosts? did you check that it work? How did you simulate non-responsive host? If power management is not configured or fail, the system cannot move the spm to another host, unless you manually confirm that the SPM host was rebooted. Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com
wrote:
Oh, fence agent works fine if I select ilo4, Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
> On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < > konrasko@gmail.com> wrote: > >> "Corner cases"? >> I tried to simulate crash of SPM server and ovirt kept trying to >> reistablished connection to the failed node. >> > > Did you configure fencing? > > >> >> >> On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote: >> >>> On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> >>> wrote: >>> >>>> On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> >>>> wrote: >>>> >>>>> >>>>> >>>>> On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> >>>>> wrote: >>>>> >>>>> On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> >>>>> wrote: >>>>> >>>>>> When you set up a storage domain, you need to specify a host to >>>>>> perform the initial storage operations, but once the SD is defined, it's >>>>>> details are in the engine database, and all the hosts get connected to it >>>>>> directly. If the first host you used to define the SD goes down, all other >>>>>> hosts will still remain connected and work. SPM is an HA service, and if >>>>>> the current SPM host goes down, SPM gets started on another host in the DC. >>>>>> In short, unless your actual NFS exporting host goes down, there is no >>>>>> outage. >>>>>> >>>>> >>>>> There is no storage outage, but if you shutdown the spm host, >>>>> the spm host >>>>> will not move to a new host until the spm host is online again, >>>>> or you confirm >>>>> manually that the spm host was rebooted. >>>>> >>>>> >>>>> In a properly configured setup the SBA should take care of that. >>>>> That's the whole point of HA services >>>>> >>>> >>>> In some cases like power loss or hardware failure, there is no >>>> way to start >>>> the spm host, and the system cannot recover automatically. >>>> >>> >>> There are always corner cases, no doubt. But in a normal >>> situation. where an SPM host goes down because of a hardware failure, it >>> gets fenced, other hosts contend for SPM and start it. No surprises there. >>> >>> >>>> >>>> Nir >>>> >>>> >>>>> >>>>> >>>>> Nir >>>>> >>>>> >>>>>> >>>>>> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >>>>>> konrasko@gmail.com> wrote: >>>>>> >>>>>>> Hi Fernando, >>>>>>> I see each host has direct connection nfs mount, but yes, if >>>>>>> main host to which I connected nfs storage going down the storage becomes >>>>>>> unavailable and all vms are down >>>>>>> >>>>>>> >>>>>>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>>>>>> fernando.frediani@upx.com> wrote: >>>>>>> >>>>>>>> Hello Konstantin. >>>>>>>> >>>>>>>> That doesn`t make much sense make a whole cluster depend on a >>>>>>>> single host. From what I know any host talk directly to NFS Storage Array >>>>>>>> or whatever other Shared Storage you have. >>>>>>>> Have you tested that host going down if that affects the >>>>>>>> other with the NFS mounted directlly in a NFS Storage array ? >>>>>>>> >>>>>>>> Fernando >>>>>>>> >>>>>>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>> konrasko@gmail.com>: >>>>>>>> >>>>>>>>> In ovirt you have to attach storage through specific host. >>>>>>>>> If host goes down storage is not available. >>>>>>>>> >>>>>>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>>>>>> fernando.frediani@upx.com> wrote: >>>>>>>>> >>>>>>>>>> Well, make it not go through host1 and dedicate a storage >>>>>>>>>> server for running NFS and make both hosts connect to it. >>>>>>>>>> In my view NFS is much easier to manage than any other type >>>>>>>>>> of storage, specially FC and iSCSI and performance is pretty much the same, >>>>>>>>>> so you won`t get better results other than management going to other type. >>>>>>>>>> >>>>>>>>>> Fernando >>>>>>>>>> >>>>>>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>>>> konrasko@gmail.com>: >>>>>>>>>> >>>>>>>>>>> Hi guys, >>>>>>>>>>> I have one nfs storage, >>>>>>>>>>> it's connected through host1. >>>>>>>>>>> host2 also has access to it, I can easily migrate >>>>>>>>>>> vms between them. >>>>>>>>>>> >>>>>>>>>>> The question is - if host1 is down - all infrastructure is >>>>>>>>>>> down, since all traffic goes through host1, >>>>>>>>>>> is there any way in oVirt to use redundant storage? >>>>>>>>>>> >>>>>>>>>>> Only glusterfs? >>>>>>>>>>> >>>>>>>>>>> Thanks >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> _______________________________________________ >>>>>>>>>>> Users mailing list >>>>>>>>>>> Users@ovirt.org >>>>>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>>>>> >>>>>>>>>>> >>>>>>>> >>>>>>> _______________________________________________ >>>>>>> Users mailing list >>>>>>> Users@ovirt.org >>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>> >>>>>>> >>>>>> _______________________________________________ >>>>>> Users mailing list >>>>>> Users@ovirt.org >>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>> >>>>> >>>>> _______________________________________________ >>>>> Users mailing list >>>>> Users@ovirt.org >>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>> >>>>> >>>>>
This is a multi-part message in MIME format. --------------5323FA587AD2C243D1E46CA3 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Should I understand then that fecing is not mandatory, just advised and there are no downtimes to the rest of the cluster if a pending manual action has to be done to confirm a host has been rebooted in order to move the SPM to another live host. Perhaps only effect is not be able to add new hosts and connect to storage or something ? Fernando On 17/04/2017 10:06, Nir Soffer wrote:
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
But actually, it didn't work well. After main SPM host went down I see this Screen Shot 2017-04-16 at 10.22.00 PM.png
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you configure power management for all hosts? did you check that it work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot move the spm to another host, unless you manually confirm that the SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Oh, fence agent works fine if I select ilo4, Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com <mailto:nsoffer@redhat.com>> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com <mailto:dyasny@redhat.com>> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com <mailto:nsoffer@redhat.com>> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>>:
Hi guys,
I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
--------------5323FA587AD2C243D1E46CA3 Content-Type: multipart/related; boundary="------------E141655F6BC3F6F8ABEBA096" --------------E141655F6BC3F6F8ABEBA096 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit <html> <head> <meta content="text/html; charset=utf-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <p>Should I understand then that fecing is not mandatory, just advised and there are no downtimes to the rest of the cluster if a pending manual action has to be done to confirm a host has been rebooted in order to move the SPM to another live host. Perhaps only effect is not be able to add new hosts and connect to storage or something ?</p> <p>Fernando<br> </p> <div class="moz-cite-prefix">On 17/04/2017 10:06, Nir Soffer wrote:<br> </div> <blockquote cite="mid:CAMRbyyuHYKMZ_KuiS_DfYWC3na5PSUvV7iumm2c-FLFKmnnsEg@mail.gmail.com" type="cite"> <div dir="ltr"> <div class="gmail_quote"> <div dir="ltr">On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <<a moz-do-not-send="true" href="mailto:konrasko@gmail.com">konrasko@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div dir="ltr">But actually, it didn't work well. After main SPM host went down I see this <div><img src="cid:part2.ECBD2DE4.FFC0D107@upx.com" alt="Screen Shot 2017-04-16 at 10.22.00 PM.png" class="" style="max-width: 100%;"></div> <div><br> 2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1'<br> </div> <div>2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5'</div> <div>2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.</div> <div><br> </div> <div>So that means only if BMC is up it's possible to automatically switch SPM host?</div> </div> </blockquote> <div><br> </div> <div>BMC?</div> <div><br> </div> <div>If your SPM is no responsive, the system will try to fence it. Did you </div> <div>configure power management for all hosts? did you check that it </div> <div>work? How did you simulate non-responsive host? </div> <div><br> </div> <div>If power management is not configured or fail, the system cannot</div> <div>move the spm to another host, unless you manually confirm that the</div> <div>SPM host was rebooted.</div> <div><br> </div> <div>Nir</div> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div dir="ltr"> <div><br> </div> <div>Thanks</div> </div> <div class="gmail_extra"><br> <div class="gmail_quote">On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <span dir="ltr"><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Oh, fence agent works fine if I select ilo4,</div> <div>Thank you for your help!</div> <div class="m_-7049131619673457077HOEnZb"> <div class="m_-7049131619673457077h5"> <div><br> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Makes sense. </div> <div>I was trying to set it up, but doesn't work with our staging hardware. </div> <div>We have old ilo100, I'll try again. </div> <div>Thanks! <div> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886h5"><br> </div> </div> </div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual</div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886h5"> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Fence agent under each node?</div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?</div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743HOEnZb"> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743h5"> <div><br> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>"Corner cases"? </div> <div>I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node. </div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>Did you configure fencing?</div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210HOEnZb"> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210h5"> <div><br> </div> <div><br> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <span><<a moz-do-not-send="true" href="mailto:nsoffer@redhat.com" target="_blank">nsoffer@redhat.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"><span> <div>On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@redhat.com" target="_blank">dyasny@redhat.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div><br> <div class="gmail_extra"><br> <div class="gmail_quote">On Apr 16, 2017 7:01 AM, "Nir Soffer" <<a moz-do-not-send="true" href="mailto:nsoffer@redhat.com" target="_blank">nsoffer@redhat.com</a>> wrote:<br type="attribution"> <blockquote class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354quoted-text"> <div>On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage. </div> </blockquote> <div><br> </div> </div> <div>There is no storage outage, but if you shutdown the spm host, the spm host</div> <div>will not move to a new host until the spm host is online again, or you confirm</div> <div>manually that the spm host was rebooted.</div> </div> </div> </blockquote> </div> </div> </div> <div><br> </div> </div> <div> <div>In a properly configured setup the SBA should take care of that. That's the whole point of HA services </div> </div> </blockquote> <div><br> </div> </span> <div>In some cases like power loss or hardware failure, there is no way to start</div> <div>the spm host, and the system cannot recover automatically.</div> </div> </div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there. </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"><span class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191HOEnZb"><font color="#888888"> <div><br> </div> <div>Nir</div> </font></span> <div> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191h5"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div><br> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"><font color="#888888"> <div><br> </div> <div>Nir</div> </font> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354elided-text"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="gmail_extra"><br> <div class="gmail_quote">On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Hi Fernando,</div> <div>I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down</div> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354m_6867735599077560094m_2305649857973554660HOEnZb"> <div class="m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354m_6867735599077560094m_2305649857973554660h5"> <div><br> </div> <div><br> <div class="gmail_quote"> <div>On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com" target="_blank">fernando.frediani@upx.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div> <div>Hello Konstantin.<br> <br> </div> That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.<br> </div> Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?<br> <br> </div> </div> <div>Fernando<br> </div> <div class="gmail_extra"><br> <div class="gmail_quote">2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span>:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>In ovirt you have to attach storage through specific host. </div> <div>If host goes down storage is not available. </div> <div><br> <div class="gmail_quote"> <div>On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com" target="_blank">fernando.frediani@upx.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div>Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.<br> </div> In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.<br> <br> </div> Fernando<br> </div> <div class="gmail_extra"><br> </div> <div class="gmail_extra"> <div class="gmail_quote">2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span>:<br> </div> </div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Hi guys, <div>I have one nfs storage,</div> <div>it's connected through host1.</div> <div>host2 also has access to it, I can easily migrate vms between them.</div> <div><br> </div> <div>The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,</div> <div>is there any way in oVirt to use redundant storage?</div> <div><br> </div> <div>Only glusterfs?</div> <div><br> </div> <div>Thanks</div> <div><br> </div> </div> <br> </blockquote> </div> </div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">_______________________________________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/mailman/listinfo/users</a><br> <br> </blockquote> </div> </div> </blockquote> </div> </div> </blockquote> </div> <br> </div> </blockquote> </div> </div> </div> </div> <br> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/mailman/listinfo/users</a><br> <br> </blockquote> </div> <br> </div> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/mailman/listinfo/users</a><br> </blockquote> </div> </div> </div> <br> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/mailman/listinfo/users</a><br> <br> </blockquote> </div> <br> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> <br> </div> </blockquote> </div> </div> </blockquote> <br> </body> </html> --------------E141655F6BC3F6F8ABEBA096 Content-Type: image/png Content-Transfer-Encoding: base64 Content-ID: <part2.ECBD2DE4.FFC0D107@upx.com> iVBORw0KGgoAAAANSUhEUgAAA8wAAACoCAYAAAA1pWe/AAAME2lDQ1BJQ0MgUHJvZmlsZQAA SImVVwdYU8kWnltSCAktEAEpoTdBepXeO9LBRkgChBIgIajYkUUF14KKBURFV0RUXAsga0HE ziLY+wMVlZV1sWBD5U0K6Pra9873zZ3/njnnzH/mnjvfDACK1qy8vGxUCYAcfoEgOtCHmZiU zCT1ARyQAAowoMliC/O8o6LCAJSx/u/y7iZAxP01S3Gsfx3/r6LM4QrZACBREKdyhOwciI8A gKuz8wQFABA6od5gdkGeGA9BrCqABAEg4mKcLsXqYpwqxZMkNrHRvhB7AUCmsliCdAAUxLyZ hex0GEdBzNGaz+HxIa6G2IOdweJAfB/iSTk5uRArkiE2Tf0uTvrfYqaOx2Sx0sexNBeJkP14 wrxs1tz/czn+t+Rki8bm0IeNmiEIihbnDNdtT1ZuqBhTIT7OT42IhFgF4gs8jsRejO9miILi ZPaDbKEvXDPAAPBTc1h+oRBrQcwQZcV5y7AtSyDxhfZoBK8gOFaGUwW50bL4aCE/OyJMFmd5 Bjd4DNdwhf4xYzZpvIBgiGGloUeKMmITpDzRjkJefATEChB3C7NiQmW+D4syfCPGbASiaDFn Q4jfpgkCoqU2mHqOcCwvzIrNkswFawHzKsiIDZL6YolcYWLYGAcO189fygHjcPlxMm4YrC6f aJlvaV52lMweq+FmB0ZL1xk7KCyMGfO9WgALTLoO2KNMVkiUbK53eQVRsVJuOArCgC/wA0wg gi0V5IJMwOsabB6Eb9KRAMACApAOuMBSphnzSJCM8OEzBhSBPyHiAuG4n49klAsKof7LuFb6 tARpktFCiUcWeApxDq6Je+BueBh8esFmizvjLmN+TMWxWYn+RD9iEDGAaDbOgw1ZZ8MmALx/ owuFPRdmJ+bCH8vhWzzCU0IP4RHhBqGXcAfEgyeSKDKrWbxiwQ/MmSAc9MJoAbLsUr/PDjeG rB1wH9wd8ofccQauCSxxe5iJN+4Jc3OA2u8Zisa5fVvLH+cTs/4+H5lewVzBQcYidfzL+I5b /RjF97s14sA+9EdLbDl2GDuPncYuYsexZsDETmEtWCd2QozHK+GJpBLGZouWcMuCcXhjNtYN 1gPWn3+YmyWbX7xewgLunALxz+CbmzdXwEvPKGB6w92Yywzms60mMW2tbZwAEO/t0q3jDUOy ZyOMS990+W0AuJRBZfo3HcsAgGNPAaC/+6YzeA3LfQ0AJ7rZIkGhVCfejgEBUIAi/Cs0gA4w AKYwH1vgCNyAF/AHISASxIIkMBOueAbIgZxng/lgCSgF5WAN2AC2gG1gJ9gD9oNDoBkcB6fB OXAZdIMb4B6si37wAgyBd2AEQRASQkPoiAaiixghFogt4ox4IP5IGBKNJCEpSDrCR0TIfGQp Uo5UIFuQHUg98ityDDmNXER6kDtIHzKAvEY+oRhKRVVRbdQYnYw6o95oKBqLzkDT0Xy0CC1B V6Gb0Fp0H9qEnkYvozfQXvQFOowBTB5jYHqYJeaM+WKRWDKWhgmwhVgZVonVYgewVvidr2G9 2CD2ESfidJyJW8LaDMLjcDaejy/EV+Jb8D14E96BX8P78CH8K4FG0CJYEFwJwYREQjphNqGU UEnYTThKOAv/m37COyKRyCCaEJ3gf5lEzCTOI64kbiU2EtuIPcTHxGESiaRBsiC5kyJJLFIB qZS0mbSPdIp0ldRP+kCWJ+uSbckB5GQyn1xMriTvJZ8kXyU/I4/IKckZybnKRcpx5ObKrZbb Jdcqd0WuX26EokwxobhTYimZlCWUTZQDlLOU+5Q38vLy+vIu8lPlefKL5TfJH5S/IN8n/5Gq QjWn+lKnU0XUVdQ6ahv1DvUNjUYzpnnRkmkFtFW0etoZ2kPaBwW6gpVCsAJHYZFClUKTwlWF l4pyikaK3oozFYsUKxUPK15RHFSSUzJW8lViKS1UqlI6pnRLaViZrmyjHKmco7xSea/yReXn KiQVYxV/FY5KicpOlTMqj+kY3YDuS2fTl9J30c/S+1WJqiaqwaqZquWq+1W7VIfUVNTs1eLV 5qhVqZ1Q62VgDGNGMCObsZpxiHGT8WmC9gTvCdwJKyYcmHB1wnv1iepe6lz1MvVG9RvqnzSY Gv4aWRprNZo1HmjimuaaUzVna9ZontUcnKg60W0ie2LZxEMT72qhWuZa0VrztHZqdWoNa+to B2rnaW/WPqM9qMPQ8dLJ1Fmvc1JnQJeu66HL012ve0r3D6Ya05uZzdzE7GAO6WnpBemJ9Hbo demN6Jvox+kX6zfqPzCgGDgbpBmsN2g3GDLUNQw3nG/YYHjXSM7I2SjDaKPReaP3xibGCcbL jJuNn5uomwSbFJk0mNw3pZl6muab1ppeNyOaOZtlmW016zZHzR3MM8yrzK9YoBaOFjyLrRY9 kwiTXCbxJ9VOumVJtfS2LLRssOyzYliFWRVbNVu9nGw4OXny2snnJ3+1drDOtt5lfc9GxSbE ptim1ea1rbkt27bK9rodzS7AbpFdi90rewt7rn2N/W0HukO4wzKHdocvjk6OAscDjgNOhk4p TtVOt5xVnaOcVzpfcCG4+Lgscjnu8tHV0bXA9ZDrX26Wbllue92eTzGZwp2ya8pjd313lvsO 914PpkeKx3aPXk89T5ZnrecjLwMvjtdur2feZt6Z3vu8X/pY+wh8jvq893X1XeDb5of5BfqV +XX5q/jH+W/xfxigH5Ae0BAwFOgQOC+wLYgQFBq0NuhWsHYwO7g+eCjEKWRBSEcoNTQmdEvo ozDzMEFYazgaHhK+Lvx+hFEEP6I5EkQGR66LfBBlEpUf9dtU4tSoqVVTn0bbRM+PPh9Dj5kV szfmXaxP7OrYe3GmcaK49njF+Onx9fHvE/wSKhJ6EycnLki8nKSZxEtqSSYlxyfvTh6e5j9t w7T+6Q7TS6ffnGEyY86MizM1Z2bPPDFLcRZr1uEUQkpCyt6Uz6xIVi1rODU4tTp1iO3L3sh+ wfHirOcMcN25Fdxnae5pFWnP093T16UPZHhmVGYM8nx5W3ivMoMyt2W+z4rMqssazU7Ibswh 56TkHOOr8LP4Hbk6uXNye/Is8krzevNd8zfkDwlCBbuFiHCGsKVAFR5zOkWmop9EfYUehVWF H2bHzz48R3kOf07nXPO5K+Y+Kwoo+mUePo89r32+3vwl8/sWeC/YsRBZmLqwfZHBopJF/YsD F+9ZQlmSteT3YuviiuK3SxOWtpZolywuefxT4E8NpQqlgtJby9yWbVuOL+ct71pht2Lziq9l nLJL5dblleWfV7JXXvrZ5udNP4+uSlvVtdpxdc0a4hr+mptrPdfuqVCuKKp4vC58XdN65vqy 9W83zNpwsdK+cttGykbRxt5NYZtaNhtuXrP585aMLTeqfKoaq7WqV1S/38rZerXGq+bANu1t 5ds+bedtv70jcEdTrXFt5U7izsKdT3fF7zr/i/Mv9bs1d5fv/lLHr+vdE72no96pvn6v1t7V DWiDqGFg3/R93fv99rccsDywo5HRWH4QHBQd/OPXlF9vHgo91H7Y+fCBI0ZHqo/Sj5Y1IU1z m4aaM5p7W5Jaeo6FHGtvdWs9+pvVb3XH9Y5XnVA7sfok5WTJydFTRaeG2/LaBk+nn37cPqv9 3pnEM9c7pnZ0nQ09e+FcwLkz573Pn7rgfuH4RdeLxy45X2q+7Hi5qdOh8+jvDr8f7XLsarri dKWl26W7tWdKz8mrnldPX/O7du568PXLNyJu9NyMu3n71vRbvbc5t5/fyb7z6m7h3ZF7i+8T 7pc9UHpQ+VDrYe0/zP7R2OvYe6LPr6/zUcyje4/Zj188ET753F/ylPa08pnus/rnts+PDwQM dP8x7Y/+F3kvRgZL/1T+s/ql6csjf3n91TmUONT/SvBq9PXKNxpv6t7av20fjhp++C7n3cj7 sg8aH/Z8dP54/lPCp2cjsz+TPm/6Yval9Wvo1/ujOaOjeSwBS3IUwGBD09IAeF0HAC0Jnh26 AaAoSO9eEkGk90UJAv8JS+9nEnEEoM4LgLjFAITBM0oNbEYQU2EvPnrHegHUzm68yUSYZmcr jUWFNxjCh9HRN9oAkFoB+CIYHR3ZOjr6ZRckeweAtnzpnU8sRHi+324lRt39L8GP8k+I32zs Bfxd5wAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAAZ1pVFh0WE1MOmNvbS5hZG9iZS54bXAAAAAA ADx4OnhtcG1ldGEgeG1sbnM6eD0iYWRvYmU6bnM6bWV0YS8iIHg6eG1wdGs9IlhNUCBDb3Jl IDUuNC4wIj4KICAgPHJkZjpSREYgeG1sbnM6cmRmPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5 LzAyLzIyLXJkZi1zeW50YXgtbnMjIj4KICAgICAgPHJkZjpEZXNjcmlwdGlvbiByZGY6YWJv dXQ9IiIKICAgICAgICAgICAgeG1sbnM6ZXhpZj0iaHR0cDovL25zLmFkb2JlLmNvbS9leGlm LzEuMC8iPgogICAgICAgICA8ZXhpZjpQaXhlbFhEaW1lbnNpb24+OTcyPC9leGlmOlBpeGVs WERpbWVuc2lvbj4KICAgICAgICAgPGV4aWY6UGl4ZWxZRGltZW5zaW9uPjE2ODwvZXhpZjpQ aXhlbFlEaW1lbnNpb24+CiAgICAgIDwvcmRmOkRlc2NyaXB0aW9uPgogICA8L3JkZjpSREY+ CjwveDp4bXBtZXRhPgqcwRHqAAAAHGlET1QAAAACAAAAAAAAAFQAAAAoAAAAVAAAAFQAAEPb KmoCuAAAQABJREFUeAHsfQl8FUW2/glb2EEQHJxxQX0JA2MQBWXIk4lIBMU8QEFcAozIIItm UASREPkLhEE0CpFFEWHQABPJyOYCwgiDI4Iw+ieKAk/FyAhCZElYL1veV91d3dV97w25N7m5 uX1P/35J3+6u9Tt1qs45daoqpgQX8cUIMAKMACPACDACjAAjwAgwAowAI8AIMAI2BGJYYbbh wQ+MACPACDACjAAjwAgwAowAI8AIMAKMgIYAK8zcEBgBRoARYAQYAUaAEWAEGAFGgBFgBBgB HwiwwuwDFH7FCDACjAAjwAgwAowAI8AIMAKMACPACMTsPXSS1zBzO2AEGAFGgBFgBBgBRoAR YAQYAUaAEWAEHAiwwuwAhB8ZAUaAEWAEGAFGgBFgBBgBRoARYAQYAYFAzNHiyN4le/9PP2qU bPHrK5mijAAj4ECA+cMBCD8yAiFEgPkthOBWgaSZvlWACCEsAtM3hOBWgaSZvlWACBFcBFaY I5h4XHRG4GII8ABxMYT4OyNQcQgwv1UcllUxJaZvVaRKxZWJ6VtxWFbFlJi+VZEqkVMmVpgj h1ZcUkYgYAR4gAgYMo7ACASNAPNb0NBFRESmb0SQKehCMn2Dhi4iIjJ9I4JMVbaQrDBXWdJw wRiB8iPAA0T5MeQUGIGyIsD8VlakIjMc0zcy6VbWUjN9y4pUZIZj+kYm3apKqVlhriqU4HIw AiFAgAeIEIDKSTICfhBgfvMDjEteM31dQkg/1WD6+gHGJa+Zvi4hZJiqwQpzmIDnbBmBykCA B4jKQJnzYAR0BJjf3N0SmL5MX3cj4O7aMf+6m76hrh0rzKFGmNNnBMKIAA8QYQSfs446BJjf 3E1ypi/T190IuLt2zL/upm+oa8cKc6gR5vQZgTAiwANEGMHnrKMOAeY3d5Oc6cv0dTcC7q4d 86+76Rvq2rHCHGqEOX1GIIwI8AARRvA566hDgPnN3SRn+jJ93Y2Au2vH/Otu+oa6dqwwhxph Tp8RCCMCPECEEXzOOuoQYH5zN8mZvkxfdyPg7tox/7qbvqGuHSvMoUaY02cEwogADxBhBJ+z jjoEmN/cTXKmL9PX3Qi4u3bMv+6mb6hrxwpzqBHm9BmBMCLAA0QYweesow4B5jd3k5zpy/R1 NwLurh3zr7vpG+rascIcaoQ5fUYgjAjwABFG8DnrqEOA+c3dJGf6Mn3djYC7a8f86276hrp2 rDCHGmFOnxEIIwI8QIQRfM466hBgfnM3yZm+TF93I+Du2jH/upu+oa4dK8yhRpjTZwTCiAAP EGEEn7OOOgSY39xNcqYv09fdCLi7dsy/7qZvqGvHCnOoEeb0GYEwIsADRBjB56yjDgHmN3eT nOnL9HU3Au6uHfOvu+kb6tqxwhxqhDl9RiCMCPAAEUbwOeuoQ4D5zd0kZ/oyfd2NgLtrx/zr bvqGunasMIcaYU6fEQgjAjxAhBF8zjrqEGB+czfJmb5MX3cj4O7aMf+6m76hrh0rzKFGmNNn BMKIAA8QYQSfs446BJjf3E1ypi/T190IuLt2zL/upm+oaxf1CnPx/t301Y5dVFjsIaou4I6l Zi3j6abWcRRbI9Twc/qMQGgRCGiAOOehwoICOl09FoUCP9S7iq5oJn77vgp/2E2nwS9m2EtO 096CA+CjWMSuQ9dd3dx3xKr6tvggfVt4imKrX7zuWhXOFZn11auEeLGX0RUtGjlqCFx/AK4a VkbI83XoiquaV14fE2jdjBpIGnvOE8p7VeWV14FgpDwGxG+yUiZtjBfnPVS7WRw1aygDGHev 9ibeN6Yrrna2N0c8fqwwBIKirzN30Pvz7Vto739Ef6F/bHZdAv0OMkdDXzKHs33Y0qtDtevF UmzDRtTQR1ct+ZfAv41+dRU1rGeLrD14Dh+kvcXo97Q+H/1Xs+htT0HR9wTGjZ/FuIGREP36 ddf6Hvf0MUCEIWqOvtQnrb3J4543inwhcKrU8c9AMSj6KhQIVF8oLiyggydMNldSws9Y8G5s LDUE73rpGl59fRWWLey1cvVT9CrMhVtpyoBu9PrX/uibQOPemEt/6hrnI4CHPn/nRXp2RwL9 PSNFjnk+wpXtlWf/Vpo56EW6bs6b1PNqH6Ne2ZLhUIyAFwIBDRAn8unu1km0w0xlFH1WkE7N zGflx+GN1LJdL+tF0lzaPu40tb0jzXiXTO/uzqU2ATbnwi/y6Olxu2jcqnS6zpfwaOVY4b++ XdKPkseu1dNtnU07P0gtlbc9u3KolVlfWZxU2vhdNl2hlt2JlRH0uQ/304D4AAGS2QR4D7Ru WvKefBoUl0TrZXk/QHlbV055A6xelQkeEL8ZpbbRRtakx0LaMztFPmn34k+zqO39mbZ3RMHx mSMRfiwjAsHQ10waQvD6NzJo0JQc85X9h2+Zw2f7sEfUntr0mEbTXxhM10ml2MG/bZ7bTO/+ 0SnPeGjFnS1opJSDytDv+cjaNa+Coa+TPo/l7KRRtzqUZs9u9KUdK7QvDedYGQzBneNlZY5/ srzB0FeLG5S+4KFc8NZYyVuyED7u/cYtp+ce7WzKG06s9ChVU7bwUR3XvopOhfnwVhrUrpvZ eZVG3UFzvqCMu64ygxR+vYqevXMgrRZvoCTsXNjHbORmoLL+OHeQVs8eTcOyVmkxpkKA7ldJ AnRZi8jhIhuBQAeI1WOb0LAlVp2nrd5PPWDEqVvHeid+7V2RRp3TLMEvbcEe+mOztXTj3UOM gCmU98VC+i30q1gIcJrzhj0J+1NhPs2ckERZ74nXKVC2FwasbNsTDPzp23fSKPkJo07d5tLn L4O3q5FX3WXKnu/yqFUXWV/5lmja6sPUA11G3br6O9+KDtHEFfvp3v+KpRrAthbyCeUVaN20 spzbTWOv7Ui5RsHM8oKmtWqEsrSRm3ag/CZqaqONWfURtOrLSXQtsK6DP3Ftyk6ih7Ly9Qfz fwqtAp9dgzC1wWchbkZmrtH6Ixj6alid2E1ZrTvSzDIA1/qJ5fTeyM5mSN/tw/zs+DEKBrt0 3WDn4N9kKMzT+8ZR9Zrok2vJaFCYh0Nh1vpdvCtDvydjuvEeDH296ZMKnsym3zZRxj0HLWRf Kpy5vGYXLwZsFRgrL1ZEX9+d46XEoDLGP1meYOhLQesLDt6ShfB3H5RLeyYka1+dWMkoVVG2 kGWLhntUKsw7/tqP7p5gzCRpVE6m4aNT6PJaRfTR0gz6aLdK+lT6Ozq/G9D5VYPLks1i1GMu bcvsQ3SBqGFTIoxDAV1OK1Lmyv105+WxVA2CdiNpJQ4oRQ7MCNgRCHSAKP5iFrXtlWEmkjh6 A83ol0B1Mc1s6cxFlDuwJY3dIIOl0tJN2fTrfXnUqY9UIPV3LSGYxTYmqn8R5rBb6aEwb1tI LZB87UugeFaScmYTfDrNhYLSh4Rc2QB1N+VLWWXc7YNaAt7oykzPzM2U0S2O6qBPqAsNZtM0 KDqznIoOUUbufuoJjag6gG1cX0k4BD8DrZtWBIeQNzlvP3WH8aR6bZS3QQgK6YIkA+U3UWUb beLQjnbrbUXDu2UsNb5UCN4HaeZVrShLw8hqa8K4tHTTQiorn7kA4rBWIRj6Ejn7S70KiQ+O og4Niba+mkWfOGr17Dv7aeBNkAXw3tY+tHAJlNjpMvrllwO0y2gravQuEzfTKw/Eoe+xG7y6 pG+mab3jNC2uKeQZ/XII9WXo92RMN96Doa83fYjihy6nJcM70yXSu93Rl5p9fxB9aVUYK4Oh vX28rNzxT5Y3GPqWR1+wGaNEIeKSKfHSA+DdfPCuLJV1z1y6n/rcHEslNmO81d9XRdnCKr37 f0WhwuwcILJpfXYqWfKfh1ZntqDxyyzii86tV6tYatIYcQfCGrvB+NYpG0J1KtWKgVANoUYV qj3FRVTs8WBdC0yINWpTw7ribqUpfjkV5jGL99N9cRUkQJ8jKsbaJ7FeRpQhNrY2ZvpQBr6i CoGABwjHwE40itZsSafLwCCmEQcuxnfDHXuHRPLh5bRthD4jUg1tvD6mk4tP4qMxrexTYT5R RIUnwR8Nm2tr775dMhDu0LqnhVACFm1cSPEwHPmMK/Mt693BCwReaOiDF2yCjxQcwdsNwdu+ 9H1LALAGNK1IXaFsT+2jzTA3qneQsqDozPRRVlNoUhVmrPMqPlIE0xx4FvjF1sP6pouxrQdY ij0YcMXGYi2jj8WMzrptgzGALiBOtSK6gP6hob98gJ1Y/3j4CFGJQU9/Cr7Z5/kpg1ZA+c+g v3iMrYs2cBEDoZm26E/90E8mHc57wPyGwpq0icfDLqv0QvGZdhcML8JodARLINorSyDMYJbC XBuGqXqyoZrtCHBVR5tocrFGZCRoxvOxLlZ8w7gm2qbPdXdmmfQfHtDYAx4XLVPyuSOI70dH 3+A7UHjeBkNfT/48apUyRilwKs15/wXq0FzSxENb3xxCw7Jl/4egvRfStmdTqDEU2z2q54vo G2EgiZfCBta771r3Ij2UrptStEwMuaRRw900EW7A0kNEKswx6KOboE3pl1MeMgyFpfR7MqYb 78HQ1+RfByDC4NXrhlidJx3jqs++X8QvQ78Y0rHSUQftUeP705Alwcnof5s1kVYAX4Ht74oL IYMacazxUg/jFwN7EhX6FDh9nfwRoL6gem9ImULW6Aw8TGfeR+MXW8b0RBi1ZtyLPv9oHrXR vNcqULaQ+fI9aARYYY4bQfOnj6cEY/CqBqHw7A95NKD3Uvp118ugSDeiLiPGU+LlBTTllo70 ji+oMbh9OT2FxCTR3k/zaMrEIbTax7qFfuNysU4hGeIGhKQVGZScNstXajTl3f30wKVQSjr2 M5WS6f/EbBRmd/QLSv3YFqbrbL+XN9PUe2A5FhfWSW1aOY+mPJFpxtU/4H/rFJr+7ETq+XvL xdz8xj9ciUDgA4T3jKgY+O+8FgYjY1ZiL9puZ6XtSkPPGfCNNcM8gpZumUQtwU+1i1bR724f qOPbOpWe63cZTZhgCXi33Uq0/mNf8I+glbsm0a/+lUk3PyLDj6C1302y1jdDEJkCt+HXjejT P9JnbfVH8MI7fngBaz/HvTHJtkeBTfCRg1spgqMpALTCoLZTH/TikfEu0mfXr4ES2OTsWig6 /fDWGvj0MA4Le8xuWjErk0bOUoRmo04EzObPfIFuAw1sF9w8c1/KoLHzVG8ZESKFpuakU79b jT4Bb2x16zqN5vQmmj5ijKqfUfdRC+n5tBRqKDNxCHlyhrlG7G76y/VSEB9F727uS5smdaQp 78mI4p5M01fNpZ4JDuHKT5nbPID1l+Ow/tLMXE/rW/SnWT770wR67OWJ9Ng91rovNfdw/Q6G 30zayIYh24pieDnzqcEDrVEz29jiVJg9tGnJZHporGNsaY1xbuZTaEMKPTCL0VITyojaDJ5L 03sSTUkZYluq1B1r66Zjbd2OJZl071jJgzq6j2VvoFE9Rbu2X9+uy6GsrDSvMfC2wdk07s+p dhr/sIpa/sF/39AOSX9hJN9mcC69m5Fsy2wHDG13m4Y2R99gC1kxD8HQ12uZy7LD1OUKqzwx mEaOiTlIs9q1ojfEa3gZpPacSCP7dda8T375UFkq4lCYY9A/lZQUUc6DLWm6nLEy2k0s+pRp N0k+JWKF2cLc369g6GvyrzPRuEm0PncEXYlxs5qjL/VSFsvUL0J5g9w3cokzI/E8gl4He/5p lMH3SdNo+8LBZl++KbsflnMY40RSNr6lWt9UDyjw2B7JY4fzKXdeFo31GpOSMb5Mso0vogSr J2E51zzxi6jfqGnU7N9jaOYG7RH/RtB7H7alHncMkS8sD6tz+TS1Y5Jp2BEB7OO4GaXcPwKn r0NhLqu+cEUsvOM8tC5NWe4gZQqjFhrvnthKf+7czfQwkTx6/uc8ukUsb6sI2SLE3mvlJkoE JRCFCrNQCJrARdJOpfiuo+hPD9xKba9ri919daGi6BfonyVGuPO76WkozP+wR9OfEudS/uI+ VOJzUxZHhOHLac/Tne0CrCPIhGVQmButsq2PtK9vtjNx96mbaQ5csMS1HmtQB/nsUK1Mxi3f Q39qpwhO1if+5TIEAh8gvD0f4tPW0KIBHaguXIzrVMOyBHhZWO7Yo2glZqAvh2J85kcozPfI ATGFVn69kFpj9vTcHrnW11Ia2wDnHQbWvX5PtPxTX8Cn0NtfLaRL1mL2+QmpSDrWN0MQGQaF WdtTAEmofOJ7zac9nxFv7KSnujbXXtoEHzm4lUVhtiepPWkC0XXYd/b/Z1KiqezbA1pC00HK aSPdbe1h1Kf5mw/TbS2MN4UbaRBmHNerARy/u03dQK8+oCs0Zt1MpcwR2HhsPXINvfdEB/3J ge1fsOlXX+EBQ3bMfaekv83ZdpgSmxkhsBZsGPaOkLTyjpdMizbnUiejjoVrMujmIY6O2hGp 9TiU91GjvI5v4XgMht9M2ngVWJ9NbAVPi+0z0a8bwqg9mKowF9HfB7ekCRvsIdSnJ3N20uPG hkSmwUcNEODvse/soUdvkmOJh9ZPauGnnDLhBJrz0Rrqbhh/rDL47htS72lDOe/sMCI7eN/m po4gDyyknVPLvwmnLKmve8D0dShKZCiztYzEa8Go1gD0FZcHs3H7TsIbQHijyQv9z6GNadRN 7q0AhXnpFrjgo781vW+Qh2o0JNl3XdhNU2/uSHlGWlIY5xlmCa73PWD6Ign//EvUM/MLeq73 VdSoBtzjldl+q+/H8pYzZe0X7XKfvfToK5Yl00O904zXCbToiw3USTNyO72ckunvX+bSjQ0R 9Bwmgq5tZxqc+2Vj8qVnHHlgyGolDVn2jMyn1iMW0jtjJL9ZZTPH9tbg6a91QzJ1zqaPn6lN t94p5QMYjPMOU88rCujVW9qR2rV1n7oB8izihuAKhr5B6Qsoe/U6HvrnGEVhlryvyhSYZGj5 BwsTyaNn9+bR73tb71UoApMtQr/cSy2b239HpcJM+2HV7mhYtX1SOAEWssH0QI++9Bu4NwpL EEFhnozBZ5mv8Jhh3pR+C+W0b0Wzle+padPo8jP5NO3VHOstrI4bYHU8+XEG3aXM0lkBiMRM ThdapczWWbNRNTC4NqpndU4inmSymFNb6albrc3M7hmeTX9ocyV59n1MYzLV2YF0WvflKLpW dJh8uRqBYAYIwmZ0WdeqbsS6UnxlA7Q9j90dOzENa5wH6IPb+f2wimKGSr8sQb7a4Ty6UbEs 2wEfSqN6vUpZy+1v9Sd9lprWpFHfZyUPWelqrqpQ3tSNqaQgUgPlvCFRdWFNoZHpyXQGm/bN XmbNyApjwJLBHbS1ZjbBRwqd6uDmKKIl7IsPydSl0wH6aJMuIMi135/NakLDF+gRe6I/oNVj aIUxEyTLevSLTOr6J4U/E0fQmKRf0a71GbRik5XpSGys9vAtEKjr+VgTmZhKw+OIZi+QOOll WvhJLiX+xuHWiXCAjRIfSqeOdfIpa540Rog4qZS3LZt+ByU31iHsy/LaZ5hFHHEl0+NP3Uw7 Xsykj/QX2n/hVvwSvF8aNbD3WeJjq3vSqXujApquljkxm/69IJWaxBbR63e2pCnGbOpv+6RT aqcO1OT8j7R0dJotj9c2HKaklvYlMUoRKvVnMPxmb3cplPjLKvrEaCOa90b8zzT7pnY0X6tJ AtpxX/ogM8PwDrD4Yd+GNPqfpxT6x6fSyG6X0QfZWYonQTLlbsulm0Ffe/vVYYpPHkUpVx+g F19X0jEQ/MPgdEo4/Bm98o7FP0IB3JzVh5rBHfwnGDg6Owwcib1TYRnLMeujJ4VNkb5BG8NY 5qsMRna4jaC8xfHU58E089XwV/fQoJsbUWMY76o7hE3BH6nX68q71jfUMKNV2I+A6eswOkk3 e1Ggmqi/r6UIRwq1bVG0MsdAMf4FdLUUZhgclq2gtqg/QSA/d6SAVs/oZvfueDCXtj2JmXgo zM9DZlmqpaTICcCFXbINUBy3gOmL+Db+daQn+sX562EEvHI3jVQMu7Iv9VKqEN9/v9iXPp7g f4Z56abBtKJTO5KcO2jWThresTnVL1lLrTQvJ6twGq/A+6dmUR61NbxMhBfUnDUbqEPj3TQd k0MyHT1WMvWEV9IKZewU73s9v5my7o/T9vfxWq9rZUc9X9hJT7fcaJNpJ+duJvpbR9vyx8S0 5ZAn9OVdIrou7yoJlfNnMPQNSl8QZQd/rh+tKMxx6bRy3mN06YXTVLORh44WbKSsXvBGVeqU +tJOGtm5OdkV5uBlC3/Lp5Qs+WcACESnwgyACj+dRzffP+aiUD02dzONwgY++gWhbwAY4J/G o1zDLB7PYz3CX1+hjd98TB9ugNKba1nRD23JpG4jpEBsrEHCZg91fsmhNsqxNNK1VSRnd2+1 FGadAezCp1SYz/6QQ7/vI4WLBBozay71ah9HtcTs3+48Gj9/OyXc0pbir0igtu3iqDYG7EtY aRZwu/YKaoAAGs6NLrQBHrOLJ7faheLJcC/sbrgXVsO6mxu7DjGwtAR5OpRH7bvJ9+LzCFq0 bjy1pALavq8xdWjdnPYsG0h9M6XiBh4x1jCL0Hve960wazMscDv0pTBXh6tTzltv0eebcmj9 11C8N8E9vJZIDZvszEqiYQsMy7fBw7Xx7dA/FNfHgBXmETR5YmMa/2ymnomI/3JnmnNLK3pL f0OT31xDO3GUnRREpNB0FLyZu2oVXGlX0fnec2l+ur7ZGGEGbTaMcLqiBAVXrG/Cpj01gWdb Bc9EbDAzY7AuZJyBm22nfhbWQjDqD2+Sw+uVuqE8wyFMDbpFn1nfA0Wnb/oso5RG/wTPgKZY A+kT21qYuUqwXD0JRsA1b42gpuhn6GQ+TeqcRCuM1GTfVOvwKkq4wzJSdhm9gaZhMzlxib6p 04NWmWe8j5n0y/NpzA1JpjDRc/RyGtm7MzUQNEQe0yfOo4atb6ar4+Op7fUdqCmMOdZGRlqy YfkXDL/ZBO7e02jyZWtp/Ku6UqrRvHMB+Ee49YtrFM1/M54GDZB4GXxWvYBmYqbmr1oY/Euc RutnDNb35nDQRAhlT8Krou4v0vNDj6S2o/x5/WiQUQbxtUs66NVb0MtDb8PNcNomPY6czaxd o4Dm3GDNVAlFYUbem5SoLSPyXqMrZt4mYuYtFq6H9p3mnX0D0XzVEC3W9qZjVgsKekGe6o6N ZRBbsrWZVyrFyGWUOuhbwPR1GJ0kP4gyOvc9kYU6cwriBNy05e7otvYhA5Vy14wK7WE44Bnm UlDy/Slg+iIZG30enEuL2u2ih0ZLeQ8BHlxO/37qKpp2YzvT7Vj2/dULV1E7nLoir4v2izAM Hljhe78PkcbWeRjbXjXGNsNwUowxu8sw2b/rOUmvsZ9WD6Se441xNw59xuLBdAaG2m6jrfCJ Q3PphcHJ2h49Z/ZupNG9e5kuxGIsX7l1EsU399AH6npdZJOauYFGdo2nPV9vodhrO9OlB9Ul W3o51P+tHs6l+SP0fMR74QnR+JKK3fk/GPqKsgStLzgwEWn5v2CwWCcMFlgWapthDl62YIXZ P9rBfIlahVmAdf6Xg7T2vaW0NMe5M7YdyueW7aEBNwrLNRTVYVCY3ze+SxcLGdwYqI//THT6 5EH66YedtOuLj+mDNdgFc5cMZCkSNYtV656lFIuQwSjMMUc30k1d1Vk1Pc/45FS6u1My3dy+ M13bQrfAy9LUxUwDZGO+XIpAsAMEqWsLgU38ULhlD06g959qQc9uMMASa7QWj9CEcjETUu+I Kvxa7TwGCt5NioI3CAL7cFhR1cufUizC+PumbXLkT2FGoxY7T8OYS0eOEh07WkB7vs2n/G1r aTVmNHfKzCUPo/xHMJNjHisVhMK8KC+ZZveRwgQ29lnQmaY/PMSY3YPSvr4/rbjNstybQpNR 1lNF0AVPeejQvgL6HmXdumkpzVcs+lLYPvJ5FiUPMRRzYwYjAQqjdsV4aPe2LVTSvBVd3rS5 6e7500dp1HOMoaordBNxYn4CfXo6FDAopnUcmwaZ5XUozKqhT6S35pkmlG5MQsoy2415MCgu 3kmJaAJnzsdSrVPbaVivXiZN9JkRopybWtF0kaB6xafQ4B7J1LHjrdTmmqtsm7HVAgYNYIgM 5xUMv9kEbhhMlnY7QH2HZujV6JpNc7oV0DApgPfOpfUDTtNtvaWQbfAZdmgf2ymJ1hmV7zkR wuqtV9GZMx6qVesUrbytHb0sgZGb9NkMWcbskpi5xOXkOXWjqR8+GEh9MgwhW/JJ0UYacHsv c3l16gtf0MjbrtITw3+xRnc26PmGfGPwXW2H94nvvgHGtGeN/Iw1vL9tUESLrm9p7Bou+6cO WurV4NEcKkNwwPR1KszGRm52t2gJiu+7rX34DmK+TYTHzAwsn9GuEswwd+AZZhOcMvwImL5I 00YftOttU5Ppbawpn2Z4iYhsM15dTgeG9qK5RhlkX3r+Pzl0Sy85yVGWfrE5HYSL/l2qi76x S75I+szueTA+jjFyGUFrtk2iH2Y0oUffMl7JG5TjTYv708rUFjTVGAyl98NWGMuGmcYya8mV jHr4s0y6Y7hlEJiMTXHvbEX08ThlNlVsFrotnYzuRIvqlGllevpdL6sML+SJxk0qVlkW+QRD X1nOoPSFABTmka/upNT2ulx0FmPy780xGUbEcsoWsg58Lx8CUa0wC+hOFUNI9aCjOVlEP323 nbZuybO7UCOMcK98JTWBLhG7ZKsMIIUFaS0W5yrPe4Vm/GWWKfx5k8dSJJwzb7ITFce21IE1 TrW8m9804dpeDimUxtTw0MfPo9Na4p2r9SaFXng7m26Pa0Q10CnFQqkI9RmwVt78q7IRCH6A sLvEEqXT0vc704y7utG/jErIAVY8Ctep2rbZIqudOxVmuXmUiBcLJUes2PsaO8H2esZQ6MQ6 PXlUDuw7P7+rKLLKt4sqzKfyaeHMLPp/f5XCtsjRcSk8bBNElPcX3yVbpIkZro3j6ZPOPpQ8 8bkrZsYyE2gSZgHl7KvK02e/X0VzXsmiBR8aswMijuOSfP4DZgX6yFkBBQ8hZDSCkIEbnTkO IwFmquRVgI2D7h1n4CuNBPhYDYqxfaZRwb4BlqHEWzPJZnkdCrN8L/Kq29RDHz5mCU6+yyxL 5fsuZ9L3woDRW3Ux9hF8+IubafDtOFsWla6JvrEO6hPOKxh+swvcC7EPRSMaCsOnaWNVKiRc GzPabKT2d9kNHPYj3ZQIvn4aXhXCUGIJZdYMrbcbsNUmhKJn81aQfPKfVdS+l1TiiVQjinQ9 Xo9Ngcx12Ea8mP+gDHBLlJevvuH0wY30wK3WrvzD4TUx6NefUfs75Ky7voxJHHkmrlC5Y4u0 A6avB+v9sXZ1tYiMSy7V8Kkwix2SPbWxC7FeDz2GQyGTL9U7Ngnr0iaZuvf+E3WBt468qmHd 7PM3WPwredGet12WkB4D4uQPf/2eTN+N94DpCxBs/Cv5Aft5tDf38/BGSvaZP64dSPc8U8r4 pEQ1+0V4QvV+2vdYWd9xlNjkNzfQvgFJtqWCepIpNGfxI7T0wV7m8pZp8Ozpgubz0UtNwL9G xnE4CWYxToLBo+gX6mN8qeVY0qjXBfGebUFj3jfiKWNMTE2MS/BkLP4SHmj/Y/G6EdK8CcVb 29sA3hXBHNNqJlTKj2DoqyZXLn1BTcj4HY9jplolJtN9fftjlt7i+5LCPOpgrvcun2wR6iMr fVTLta+iT2HGZjlTJrxBe09g7c+GfJqDjXTugFnrxAm4QZzDjpOC1Ee30qSu3UzBVg4idWpj Ef/TljAo3+uDSxEtwZq7CV872krnFOq6cZVp+VfPzXQqErITFW4UdQ4EozBjbRIs758smkeT 4BrqS+DSS6ds+uAoLj+6C4HyDBA2QcALFvuMVH14KpDt7EBFyHbMMMt2LpQ1eU6lPS8rrlCK 968KQmE+vZWe/G9rPb8s/h/uS6V/vi2FDbyVAg4ExEJY7u+UlnvlvT/B0b7+EmXGZjzHXk2i QdLdW2aKu3BBzejmoTHtO5oCisThJGYFOpuzAlak5L4ptHapJUxJgdeuMFuKTiyEkvrWmEvH YQwU2nNNCCx7VaODrBs+l0a32lCYJwWgMOvnM9sFcLPMK6HkT7TqYtXS+5c0xIj2cSg/h2ZN TKPlymyNM8ZYbGL4aBXZxDAYfrO1fWOGan77lj6EXCiGYgkEtnFqb24Io/PKr/fmwBXfmqly YmR7NuhPjk36pJGqFoyoBz7wzXPCFfpnlR9lWgVwt7zXEoalwiyE5SaIIy5fCjP8qm3xJE+o fYPm2aUYquPhuvn0ddj0zlxGkI7ZrFH6bBbaeyhd8wOmLzZVysKmStaxcvrM26UwPFjriAU6 4BvjyMo2Sal03z39qVePDtQQ4WztAx4li7AmVhy5Z16os/PSZujgITIO62ZzjY+Sr0pVmKWi wwqzE1K/zzb6yL4V+O1ZnER9s3wbQGU7//E9KMwTAusXC7Cnx73pcgyzj5XiWDkbn6mlfmgu zbh6I/05U8ZVP8JbbJvwFvPQmiktKP0d45uiMNeBnFxXHAzuR2Fem96CnlljxJM44FF6MdrH SyOcejM8ny6DLAEROCRXwPxbkfqCcHl/y1gmI2vng3eFh0xdKMzWhFnwsgW7ZEugK+YedQqz 8+zj257bQPP/mGCiefYMZmdOF9GcPi1pltQ4DeaPrY4Z3PGKwqwMLrGHcuh6ZT3ygIzl9FBy Z2oqBraDsL7fJa3vVgdXmsLcGOvL1N3zpOVdZwD77J8USs2BEOc/f/dtEZ33/Ezbv/yM8jeu pZx1ykYtKFIGznzu0xbnaUI44su9CAQ8QKhQOM9bVr8p6/dlu7MPiP7bubQkqxt62IQOZdZU KMyH31ePsbIURG2G+Zh9EzLfgkgCTXhtLnW7QV/PfwjrebuNNoQGObBXkMJ86bYsum1EpoqU 9nsydgTtfsVuGoPNVD4yvuplJfrwiRY07mPjZasR9Grm49T+NzD1YzD9KBPW/mX6N8nnh5HH HUOtPGaswU7UEGY0PGoKhbUbdj29lXreHE9tfp9CneIb2YVupc7CGHDBj6EjYIUZUk7j+r4V 5uNfz6KkAYabMaozAycBiDKju9WuY8fhj14rFi7kteFCrGv9Wrtq4KED3xVR0blT9P2OrfTl F9i0bbFdwEzE+tpX7kugRmgrQp4L5xUMv9naPmgjzsi2rbM3K2S0faxva99HKqcGn3lgILqt G200wg566Qsanvgrgke2dnng738GvhwN6gBfcbA32tYZm7Jq8WttrBvcv9K3wuxlwJJtCTPF 7ZWZYumSrRtRRBEcO/UaY6dTYfbVN4jYxaWcQCEVQRFO7VPEc0VfwdB307QknMphKU7C7XzE bc3NY/q0MjqWwBBNou0FI7Sjf2ztQ+kbfdUtBv1YDZC3oVii4ZjdJqPPjq2hKOsOl3EZhmeY r/QFr893NvpIfhAGB2yelQ6Dhdx0TY0sx6mTu2fBWBpYv/jD6jR4GEmlV+Fb9H9CYfbHK73g nTLmyg99Gtak54Mw3KyFl+IzZqEtl+wGUGRhw6TCj3HEXariko1NartfjXEMCjMOLdAvRTaW Bme7fIBgndJp8m1HaXzmLCMS/LTAG091a06NQiSXBsq/IdEXzNraf4jj5WphDK0PncGOFWgM Y3zgsgVOtNDGZHs+/BQ8AlGnMJNj10oB3aDMNTSoZwf6tRhkiuFWvWQ0DZtiCWVmZ4L1gRvg dvLUewbgUmm4ALfS43l0vbnbIKzAm2AFFr0LLvuaEKuD81KYocT2jEMjR7zG9ewb7vTE2qeM u+B6CAY4++UsbFhmdbJSkN63NoP+5xnZ+WAdzZZcukZasA6upYfu6mfOOssO2zkrpZeY/7sF gUAHCHu9rVkP+3t9YBO7OYpLzEg1QLv06uQNt2p7O7fav9qZ24QOCIXapl8Q/GpDqaqHmTB5 XqzITwrVQnjfPkc5XxLfZLv+cW0aXN0MoUJbrzVYG+zF5nzz+2M3ezljqQg4FTHD3BIKfHuv fQQMRQdbU0+GwrxcVAKXqTArgoa66zjOk6GHulmuuZLPY7Ab+U3mbuRICOte14/FTsVQfs9j A622pisXUXec0T7nnriwK8wxh1ZhHbs0GhLZNrdRNiqLj0uh4c/PpUScEmDNohJNfHs/9bgm VvcAAg1nYzO1+RqK1mZooVy7amR10Vsw/GZr+4ageQzrBLsp6wS1jPFt29Q+jv0tDH5ybvqF tc/rp6bqm36dgadAp46ax5RwAbx96DQa1Bnrm/0pzOCr0rw6bN8k/1wooNk3y528RWn1Tb86 t4qlxjB6bHptCD2kjKlSofZXBrVv0OruVOy0l+Kf3dNFzmaZnyv4RzD0dc7IiSKlTlxDjz/U gUQPuvfrPJrgOP9aW8v9h+ZUHcrtPrjg9njSW0HSxm4oSBA/tKual7WoiN50eL11GbqQxj50 B/3XlTjyrljssD2QRs6zlHmz/xEKH/oTJB9VVzD0tfGv5AfgJzZ1O+3H0CPHqerF2PSrSwD9 Is73dSrMtrFSEOxEPg1rbW2YKAmoeadclk9/xl4Hn8iXxl1OyIjH09sz6L8fkXIk+ldj06+m zeCBuX8jjezYy1xioG36tWUSjpXEUhxlHJNjlboBn10+EOMfjpW61nGGuFjatCmbrv+1vlTL UcxyPwZM3wrVF7ARKIyhQi2oD7lG5S0n79qx0hXmwGULVpjL3WAcCUSfwgwAnDsAOzDxehz5 Jo6raI3FlFCYP32xBT2+xBEE5zBvm3Aaa6oUl7hOo7D5xq30w4fP0svvWAOS6pJdG1v6/+72 IY7EiCYs309/bI2NXpS1TyJQl4cn0Q01P6OX5lrKvPYeu+dOw+65x+HamWRz7Uyh8a+k0rUl B2jl22m07F8yK2s3PmsGQH7ju5sQCHiAcFS+cB2syV7nCMMgtNFwCzQEAzEIeHXyF1GYa0DJ lpbkvSvUWWRZCAwUOxbS9T/lUCuVtyAkDxqdSjV2jKG578uw+l0KIk5Xt8SH52JDDWxI9coQ Wm5scqLFUASccivMWn0P0pvY1TdbLZa0tuNoukwozMaEsaUwqzPMiDdoYi51qL6bZqRn2PZC kEJITI0iWjuypXIWtsgsmXrfR7TsbdWTBEeabMilG9B1/YhZCXONuFLnypph1sqMjXDGfm0B E987nf54A9FbEzLNzaJEPRZvzaWbqm+lR2/sRhus4DQ0fS7dAqX5q/dzKOvvVj3lrsDS00GJ Uuk/g+E3nwI3vDsG3GFtoiUq0jMTRlOc2GDfPMcyQO1Z2Y/6TrRwobhUenZgAn2zaAwtVXAf nbuHBmMTy3NQmNv42NXeaxZZmdX0+qa0pcJPsKQhTSp2OvSJ90Jp/yaHViv5W0K2/1lutW+Q RPw8O4nudbq4yvwRqDLoHwx9RfnLcia8rKfYCX0NzrbXdp3HywMfQ2GWS0UctBAziqVdhetw lrmi/JQWVnwzTz1Q+vWLxXHT92Do65N/TYMDjM7qRrEGWHKcql6niNb0DaBfFN4f8LhKSrMU Wj1Jfaxsr83MIs+xzr1ssJu1odiuGQmXa1MWFLGtJQ3Clb+JmKxRzozW00+me+8n+vvflP4F H3pic78MbXM/zEwrLtlyrCpdYcYEEc5jP7Mby0mUo+OE0SYbG4w2wbhV0Vcw9C2PvmCfYDMU ZrNt+K+db1kqUNmCFWb/CAf3JSoVZuEe9uaAVjThnxcHTZ0JEZ3JCQxenR1CAcWn0yerUmgB Opl5F08SGyrsofuw6ZbT5VFGHTEfbim3N6d/zUii/i+pyrYMYb/Lzknb9Aud1sj37N+9nuQu qfjgZcn3CswvIhmBYAYIW31hrR4Ea/V69aVxtIt4FQNNWQ5szk5e7qzrb4ZZCN+mwOc4U1XP LoHmbdpAtzc/SK9c14pe0l+W+l8KIuexAdktd3sbo7wjG7PZ9cq/hlnWd/sbSfTIHItvpaIj znJ/BgqzFDlkWX+Ei/g90kXcu4DWG3i0rM9O1dY1Njm3kXq0tytUVkD9l+kZg0fbrIRUMoyB uzSX7KewhlluWCTLW70W6oFjpbzea+5fvl2yNWUGZX4YZd7gLKjybGKFd3vXYXObsXYFTAlq /ISwtwXrV9E3V4bC5J2//U0w/OZb4D5Ii7Cr9MtK8nIWyDkrK9td7QYHafGjrWjKBiWS8yd4 dxOOZQLb0jls+tXR3InVOE4MH3SlGEfXPCENs85viru2rS156H2xx8ffnJmqzwlwx19DiZgp E9dZ26ZflvJv6xtkdHgiqJ4m4rW2o7pxPJr0dJHBQ3EPhr56OTz0Lxjr+r8oMfVXugSalreG uhgbmGnr+DHDbO7eL7xvYJgT3ms+MfJK1pfy5BVIe5E4Gjts9+ug/dYUpya+w7n5bTD09c2/ ygw91sE6+2qzLxV95qnA+sWaWK7XtvsQBxmMsbIFPA7gaXDIaeiWZ3MjlrNfNb0K8E0uafDs whraO5x52LMUewksMI+B8tg2/bJkUsv93y4fSA8rodR5aPPkFjTMnIhCH/E+xv04RT6wZx30 UzD0DV5fcCzhtPWV9hlmZ4XsWFk8H6hswfK9E9nyPUepwixAw9mQy5bSqyPTzDWFKpTxnUZQ /wGDqXv7q8zX1Wpjk6JaBfT6kHYOoUR3I2l5fCu9PqkbvfaxGQU/Uugvix6jg7O70cuGH4w8 01IMeIc3Yw1LquVeLWJKobF6rSL65+whNNLc4l98TaAnps+lDueX0oOj9HUk0l1bFxg99MXy tyjrz2O83G5EWTJmp2NtI3oi46oKR7HIsvC94hEIboCwl2P9JGxkpbjtibN9U6/Xzb+xuNWv pYe3d/I4RgkW7ZZQZGoXrYInhXQ7M3gFcewCn7fbpkhVKgmxJfk0+8kkmqPyFmbQsjMfozPL h9BTi3QFVYYXA8XZ7/PoKbg6ynWdWimTJ9G8gVfS4tSBJt+L9fw942PJNkMGN+dN4jzkUqzB /up7fGcWJaVm6qCIOmC9bnehIOBc1Odvtta0yfeirD99mEl3P2GtCxORuwxfSA+3+pH6p8n+ AQMnzqdu1dAQQk7spr9mZtBzi6QKbmSJo5cmPzHR1nf9tBHHSkm3TmfdbOsnFfo4Nv0ysYXC rJ7DbL7XFGai1dgNeZhhOZTrS01lFmV+LX0ITV1mGRS0Uvsos3A5PfXVKsp+cSDlbDLqptwe Gb+cUlNwNjPamLgi1iV7BRQiaYSVxii0u68X96MBWZK2Fj/ZlUzrvc5PRfSP2S/S4OedM1AJ NHziC5R6Vwd9aQLwOrcPCrO5ay3obpxjLNI5+RE8S4bI9mjPQ8xwJcsZLuk9YfKJh/694i2a PXMMfSSXPWjUgWvnw9k08uFUaqlsWFUHLqltTJdUpe2hDKYxzYhP5HQx1vlB2wAL+fs719iM XgE/ytuf7vl0Lc3P7uezPaeOXkj9e6dQU6M/FcWt2xR9g9hkTbYPnHtr9qs+MfJVSQ99/rd5 9NLTGT5kAoRHPzr5yVG2/kIqTr5Sc/O7YOhr845y9q0GWDalGu+cfSYF1C96KP+v9iUOIhuZ ptiYr/6prTToemvTS/W4tmqOo9zkEieRhrqkwbMvnxbMzaLnFziNPMk4Jmsa9VRkY7G532ev lNLvi8Rt4wzRNGxg2OUKY9IGRgPVqCCMvdkPYpa5go02wdBXFD04fcFDGzOVmX5prDT7Sj1l X/8rUrbgXbJ9IRzcuyhWmHXAjv4ijpY6qB3BUqsO3EOwC02DBo2w8YwdUGFxVc+FO7IXR0Cc wBmX1RtRg8a6tVzEqIvOquiHg1R8Hms7a+FbA+ubPUV0FFDAG4t109ikq2BPEVR4nEmKvBsY eUsB8NjPRbT/KHZvqY5NW0TZDAHRmZ4plOJDEep1sriIjsldX1DOpko5tbiwRF6CAdlr6ZMz YX6OWASCHyAcVQZfHMa+TCXqa3T8jbFOy9kcz57AVgAnrYDaAI4B1fnerjDr4c8XFtF3h7zb ujZ7BFY68AM2gNLOlr0IbxnKm8jzp70Htc2latVBnLr++VEq/yePoE/AjvnaVQZB3FkvWd9j h+ByKhcYIjFzx2CkfRh5SCylFfgk6v0f1F+s3vLVB+kFEglBIALuoibncXTU0YNFdAjnN4vL 2R+Jd+psUWl1c9bDpI+f8pK/9yJTXKIPOmdUUu2bRJkL9xdRsSgzGo+vftLECuloZUZfJuuo 4dPYux8UykWdMHdm5eE3f7Q5UmitUxW4SmXGSS/Z7sSi1iMYMwqP6PiKcaMpxg0boypCm990 JPaSF/Bs5uH85uATbfzBMUnHZLtsgDPBHWOqPG4KSZWpb6BiuxJAyqyZfVdtkWJorvLQV5bo 2GGi40cNbETn6WdcV/ERcW3tA88qLcT3i12SJh70n2cgnwg+UuUNGd8p68j30XAvD31t9HHw g8DuAvq9I8ftKMq+X7wNtF8sdazE+Ccu7PVHJzF2mxf6x8boJ0Wzs5UXz+o4IcNrxxMe99Cx Y6fRZvRxuSn6XtuFNNVjoPz1+zKOs79RMSjGmHFWDoyIUBOeXw0VA5tMI9h7eegr8gxWX1Ax EeN3WY17TqwkzwcqWwSLF8ezIxD1CrOAQ2zQKhRlhU8tlNC4a2CgF+fI2a6z6PwQ74IjkhAy a0PAKIag7vwmdq+8gHzke1WIxDHQdErt2JCZKTRCACqCgH1OEb5FWYSQUBP5e1AWcanpiefj R1EvfHMUUXzSNhZzylDaB/7nKgTKO0CoYBRDATyrtEFp0FHDiN/OTl6ei3oePHFUUaRNhcyR gBAodYHO+iCVhAunwQsQOiQPyRDiDF5YnMzyyfDi+6lj+EM8Gx+Ar4VR7CziyPfSgHUSfHPK 4Cl1DZbMy3n3V195ZqMMb5bJoWia7xHwOPjcoygoIq4QZISRzKP0D1K5F99Fn1KMOp5XaCPe i0v0OQ3Qd0k9srS6+aWPv/L6e69nTWp7cfZNwoJxtBhlluAbcWSZnf2tRkNBKx/hq9WAoRKG R7H7b7iv8vCbkzbSGOXkB8lP/tqdxMAZT74XeNW7BG3KeOFMR+ySXc/A8hR44aTkBYSXeYuo zvI6N4nyN66K3WBrg151ZQGQlv+256EdX+7CgHWUVjzdi15X1kKPwd4i94m9RXCp/KC9CNG/ 8tBXLVJp7Vnwex3gUwdGRvWy4Y0PKi3UcKX9Lk0mEPFEf+HkvdLSc9u38tDXRh+ML05+EFid QP9+Wunf1b5fwzLAftEXj9vSRJ95GP2s7DZVw5KT733tGyDKdBaK/gmM277GF9h6qLFDNi61 30d6Tl5XDUNOo4IvJV7DKch/5aGvzNJfv6Z996MvOBVmX21Dpq/enTSSPB+MbKGmy7+DQ4AV ZgU3MRl7HpbXahiwLkD4FDtU1nIMWkpw7acWB2GrCUbBjDGimtcZCOlCIBQ74NVExyKFVjOA 8wc6UqH8CrlXCDWxjrzP49sZ0dkaTFnzognqGcgyak+IG4tyljGqngD/j1gEKmKACEflz0Nw 0JRm0dbBCzUNAV6WRSi6woBUZt5CRA8Gfo23wKSxirAu06wyd/C4VNjL0gfJcpuY4YXoj2pG AJ9rdBSzXShvWWip1lHUWwj4Ze0HRfhQX1WN3y5gzNAMqsBXYCyMRNUrufMXRzWe02Y0QWMf 41ppNHEe62KFlefG4g3qU1meUhVNX7X9BzquW1gE/kvQRMg40vAYKF0CzzEyYlQ0fYOttdou LtYvmn0i+NvXWBlsGbziCfkUf1qbEX2JQ971Cl8FX1QkfYPRF6ogJFykABBghTkAsDgoIxBp CFTkABFpdefyMgKVjQDzWwUjDjfsu7EWc4ctWfvGYdIzxBYkRA9M3xABW0WSZfpWEUKEqBhM 3xABGyXJssIcJYTmakYnAjxARCfdudbhQYD5raJxh0v2ug9pF9bpHz1emy69Io4S2iWYG72J 3Oo3g8dIRWfrJz2mrx9gXPKa6esSQvqpBtPXDzD8ukwIsMJcJpg4ECMQmQjwABGZdONSRyYC zG8hohuWYBzCWkznpa5/dH4LxTPTNxSoVp00mb5VhxahKAnTNxSoRk+arDBHD625plGIAA8Q UUh0rnLYEGB+Cx30YrMdsfGQ2MBI7DMSi91zK3ujN6Zv6OhbFVJm+lYFKoSuDEzf0GEbDSmz whwNVOY6Ri0CPEBELem54mFAgPktDKBXYpZM30oEOwxZMX3DAHolZsn0rUSwXZgVK8wuJCpX iRGQCPAAIZHgOyMQegSY30KPcThzYPqGE/3Q5830DT3G4cyB6RtO9CM/b1aYI5+GXANGwC8C PED4hYY/MAIVjgDzW4VDWqUSZPpWKXJUeGGYvhUOaZVKkOlbpcgRcYVhhTniSMYFZgTKjgAP EGXHikMyAuVFgPmtvAhW7fhM36pNn/KWjulbXgSrdnymb9WmT1UvHSvMVZ1CXD5GoBwI8ABR DvA4KiMQIALMbwECFmHBmb4RRrAAi8v0DRCwCAvO9I0wglWx4rLCXMUIwsVhBCoSAR4gKhJN TosRKB0B5rfS8Yn0r0zfSKdg6eVn+paOT6R/ZfpGOgXDW35WmMOLP+fOCIQUAR4gQgovJ84I 2BBgfrPB4boHpq/rSGqrENPXBofrHpi+riNppVaIFeZKhZszYwQqFwEeICoXb84tuhFgfnM3 /Zm+TF93I+Du2jH/upu+oa4dK8yhRpjTZwTCiAAPEGEEn7OOOgSY39xNcqYv09fdCLi7dsy/ 7qZvqGvHCnOoEeb0GYEwIsADRBjB56yjDgHmN3eTnOnL9HU3Au6uHfOvu+kb6tqxwhxqhDl9 RiCMCPAAEUbwOeuoQ4D5zd0kZ/oyfd2NgLtrx/zrbvqGunasMIcaYU6fEQgjAjxAhBF8zjrq EGB+czfJmb5MX3cj4O7aMf+6m76hrh0rzKFGmNNnBMKIAA8QYQSfs446BJjf3E1ypi/T190I uLt2zL/upm+oa8cKc6gR5vQZgTAiwANEGMHnrKMOAeY3d5Oc6cv0dTcC7q4d86+76Rvq2rHC HGqEOX1GIIwI8AARRvA566hDgPnN3SRn+jJ93Y2Au2vH/Otu+oa6djEluEKdCafPCDACjAAj wAgwAowAI8AIMAKMACPACEQaAqwwRxrFuLyMACPACDACjAAjwAgwAowAI8AIMAKVggArzJUC M2fCCDACjAAjwAgwAowAI8AIMAKMACMQaQiwwhxpFOPyMgKMACPACDACjAAjwAgwAowAI8AI VAoCrDBXCsycCSPACDACjAAjwAgwAowAI8AIMAKMQKQhwApzpFGMy8sIMAKMACPACDACjAAj wAgwAowAI1ApCLDCXCkwcyaMACPACDACjAAjwAgwAowAI8AIMAKRhgArzJFGMS4vI8AIMAKM ACPACDACjAAjwAgwAoxApSDACnOlwMyZMAKMACPACDACjAAjwAgwAowAI8AIRBoCrDBHGsW4 vIwAI8AIMAKMACPACDACjAAjwAgwApWCACvMlQIzZ8IIMAKMACPACDACjAAjwAgwAowAIxBp CLDCHGkU4/IyAowAI8AIMAKMACPACDACjAAjwAhUCgKsMFcKzJwJI8AIMAKMACMQWgSWLl0a 2gyqWOp9+/atYiUKbXGYvqHFN9ypM33DTQHOnxHwjwArzP6x4S+MACPACDACjEDEIMACd8SQ KqiCMn2Dgi1iIjF9I4ZUXNAoRIAV5igkOleZEWAEqjYC+/fvpwsXLmiFLCkpod/85jdVu8Bc uiqBgBS41x1Zp5XntSGvVYlyVXQhZD2jdYbZ7fVm+rrbcyLS6HvkyBES43BMTAxdcsklF+3O jh07RmfPnqXq1atTo0aNLhrejQHOnz9PRUVFWtUEZgK7SL9YYY50CnL5GQFGwDUI/Pzzz5SW lkZSoJAV69ixIy1ZsoSuvvpq+YrvjIAXArLdsMLsBY0rXlX9SNYAAAObSURBVEj6ssLsCnJ6 VYLp6wVJ2F+cOHGCEhIS6Pvvv9fKsn37du25tIINHjyY3njjDbr11ltp/fr1muJcWng3fsvL yyPZT23dupXat28f+dWE1SQqr8OHD5ccOnSoBJagMtUflhIt/NGjR8sUngMxAowAIxAIAhiQ S5o3b16CUcXnX4MGDUp27NgRSJIcNsoQePvtt0vE35DXhmh/Za3+vn37SnJyckr69+9va4P3 339/ycaNG/0ms3nz5pLExESzvT799NMl/sbILVu2lDzzzDMl7777rt/0yvpB1rOs4d0Srjz1 hsdKCYxuNvreeOONJWvXrvULD9PXLzQh+cD0DQms5Ur01KlTJVCYzT4uLi6u5OTJk6Wm+fjj j2vh77vvvhLMtJYa1q0fYfwxMYPC7IpqCjeDqLuOHz9ecs0112jETElJuWiDFgzTunVrLTws RiXnzp2LOsy4wowAIxA6BEQf06FDB3OAGT16dElhYWGJUGTGjRtnvhf9VnFxcegKwilHNAJS 4A5UYRaCnT9DjXg/atQor3FSKFoyjjDmyN+ijToFyjNnzpjt+x//+Ee5MZb1LHdCEZZAsPWG e2jJ0KFDTRpJWsn7Sy+95IUE09cLkpC/YPqGHOKAM3AqzIJnsrKySk1HKsxl0S9KTSiCP7LC HMHEU4uuMkBZLECBhlfz4t+MACPACFwMgXXr1pnC7KBBg7yUk4yMDPP7ypUrL5Ycf49SBKTA HajC/Mgjj2jta8KECdqM8v/+7/9qM4/SsCyExA8++MBEVcyaJCUlaXGEsiVmLw8ePGgalp1t dMWKFVpYIUBWhMFZ1tMsUJT8CLbeCxYsMPuPKVOmlBQUFJTs3LmzRNJd0Ff1JGD6hqdBMX3D g3tpuaryvzQwiTtcs/1GY4W5pIQVZr/NI7I+qAxQFgtQoOEjCw0uLSPACIQbATGjLAZhMVMn XLOd1969e02BVwi5QkHhixFwIiAF7kAVZuEmLZYoOS+xdEl6V82cOdP8LLy0xHuxhEAoyvKS ipkaVh0/K8o1T9ZT5hst92Dqrc7ujxkzxgaVUIx79Oih9S2PPfaY2a8wfW0wVdoD07fSoC5z Rmr/pSrMpblml0VhFobDH3/8sUQYJ7/99ltt3BeeIP4u8U1+Fx48X331Vck333xjW1YqvqsG SWEUy8/P18L98ssvXknv3r1b+y7CYP8Ur+/OF8K7TZRVprlnzx6/S3DcqDD/HwAAAP//guTa mAAAMLtJREFU7Z0JvFVT+8dXhvJveMtMMr5es4yJEhlCevGiKN5IVEpFUUKh0iiKBlNJSFQa 36REJEqTSgPJHImXJpSK89+/1ftsz1l373POPffce/e557c/n3v3sNZee63vs87a61nrWc82 sRzctmzZEqtatWrMGBO77LLLYn/88UdCCvmNnzAxBpIACZCAIqDbl5o1a8a2b9+uQnce/vnn n7HGjRvbNmu//faL/fzzz3ni8AIJjB49Ooa/5k81t3+ZIHL33Xfbejdo0CA/Oamzbl184okn bNxnnnnGj4v84F3bsGHDpO9a/6YkB1LOJNFKXHA65f7iiy8sf8hg+fLleZhMmTLFhlepUiW2 adMmG0755sFUJBco3yLBnK+HyG8Bv5/OnTvH6tSp4/+eHn300cC02rRpY+OE6ReTJ0+Ooe1E mvoP1yZMmJAnzdWrV9t4FSpUiI0bN87XX+TeOXPmxMaPH2/jHHHEEbGPPvooVrdu3bi0Effm m2+2/YsffvghrhySTseOHWNbt27N8/zPP/88JmWSuHrftGnT2Lp16+LuGzNmjP/8+fPnx4Vl 64nJ1owXJN/6BxBWoXX6yeLrDu5vv/0WW7x4cWzp0qW20q5fv14nxWMSIAESiCOwcePG2HHH HWdfLj169IgL0yfywkLHFvdwIwGXgHS4M6Uwb9u2LVatWjVbN4MUZrcuSidJ4v7yyy9+3V62 bJmb3bTPpZxpJ5ClN6ZT7kWLFln5uYMbguDTTz/1w9GRxiZ9HspXKBXNnvItGs75eYr8FqAg Pv/887EVK1bY34sojEuWLMmTnLyrXf0CA9/t27ePu1/S0XsooFqvWLVqVdw90l+QexYuXBiT tleuhe3RDgQp6xK/Q4cOceWBLiNhifa1atWKy7PODxXmOKTZdaJ/AG6FDipJovgy+o4K/t57 7wVWLIxC7dixIyhpXiMBEshxAitXrvTbDT0z52LRL6BPPvnEDeY5CdjZZXS6M6Uwi7KFjtLU qVN9wloR1rOW8j4UhXnYsGG2bmNmA51FbLCOWLNmjf2Ta37CKR6ko1ikmHSko6VT7pEjR1oZ uMqvFFR3xqVjS/kKnaLdU75FyzuVp+n+v7RrvXr18t/ZRx11VAwDi3oLU5iHDx/u34c2FfLG JBt+b6+++mpcmDwL6erfKO7DTPPQoUNj999/fwzPh3Kt+weIg4FOTN5h0u65556LSxvhF198 sQ3/6aef4sJ1O4H2+YorrvDv7dmzZ+zrr7+2s9AbNmyITZo0yQ9DmtJ+IM86P/o6wrJ14wxz AU2yW7VqFVdhUGmC/q655pq40ZdsrTDMNwmQQGYJ6Jehfkm6TymJLyC3jDwvGAHpcGdCYf7x xx9jMO/D+wx7dOz0Jp1CdMzmzZsX051BKNHoUMn9n332mX3/derUKe79iM4ZBozyu0k583tf tsdPp9zSbqAPErT8TCsE6GDLRvkKiaLbU75FxzrVJ+nfh7yfoeDKsk60j65ptvx29IQcBgql PcQ9CxYsyJMFPZuLWeC1a9faOLqPgHsxoyybTMbJ7xzhUKLd9loGMxGOpV+uko+ZZYRBGYcJ Njb93LD2A3UW9+FPK8Y6P/q65Dsb91SYC6gwyw9DKkyDBg18W36YaugfSKLZo2ysPMwzCZBA wQnol5K8kINSLYkvoKBy8lr6BKTDXVCF+ddff/VNsfFu07PIkjuY77qmgYgrnccBAwbYTpSY +D3wwAN+h6x3796x2rVr23MozbJ2VtJOtpdyJotX0sLTKbe0G2EdXq0QaDlTvkVfeyjfomee 7In696HfzxgklH4/9to0W/QCrTBrSzKEh23STiJNUTR1H0Fb6+g05HeO+15++WUdZI9ljTPC 33zzzYTh8lws/Zo+fXps8ODBsQ8//DDPPbigyyX34brOj76OsGzdqDBnUGGuV69enllk7XAD yvPmzZuzta4w3yRAAoVAQL8M9QvZfVRJfAG5ZeR5wQhIh7sgCjOUZcxAoGOFv2nTpoVmCqaA 6HyNGjXKdtJkqQBmpzFDgtkKmPCh0ynK9dy5c216uCbro7W5d+jDVICUU13KicN0yo17IEd3 jaEA++abb3xZw3mQ3ihfTaPwjynfwmec3yeEKcxIRyu32jQ7SGHW729pA4PyAsVU2l60q9h0 HyFs4k2nry1F5Bk6XM9QB4WHKbgw3/74449jb7zxhlWimzdv7ucVedb36efp6/K8bNxTYc6g whxWKbp06eJXqrA42Vh5mGcSIIGCE9AjtGEvQzxFjxDrmaCC54AplBQC0uFOV2GGoqtNDV9/ /fW00MgaP6yxwyZ1XK+Pw3XMRqOjJbPSuJbKJuVMJW5JipNOud9++23LGAMYQd71xemXNsVM xozyTUYovXDKNz1uhXlXIoU5zDQ7mcKcSA/QyrEMoOtrYX2EZApquuFYxzx27NiEjsJEwdfl Sva8wpRZYaVNhTlDCjNmj901AyK0iRMn+gpz0MiPxOOeBEgg9wjotU3oiIZt0kl1lY6w+Lye ewSkw52OwgxrKL2E6N13300LIBx6oQMVtAZPmygi8ZdeesnGhcOb/GxSzvzcUxLiplNu6WzL bL/LQZyVhinUbnzK1yWSuXPKN3MsM5VSIoUZz3BNszEDK74adHuXqgIpA1hoQ4MUZq2U6jIm Sz+dcFiYaKdfohhjD6dhffv2tY7H5LrOW7Ln6bxnyzEV5gwpzGHrg1AR5IWFSjVixIhsqRvM JwmQQBEQ0C/kMLNJOOuRNZ+pdmyLIOt8RMQISIc7vwozTASl04P6hRnhdDdxLqNnjfUMs16v 3L9/f/tcHTeV50o5U4lbkuKkU27t8Tqo/yHyQudeHAglYibxtcwo30TEUg+jfFNnVVQx9ftZ FFj32do0+9RTT/WXtIQpzO7SB52eDGChPZaBRK1DaKVU35dMQU0n/LXXXvPfCxhwe+WVV6wj MjCRDZ8LlHeHNvVO9jy5P5v2Oa8wN2zYMNBzpBai/sG4irGYXrjX9f36u235Xaul0+ExCZBA ySQg7QhePPAo7G7fffed/1JKpc1y7+d5bhCQDnd+FGbtZwP1D50kzCJipkP/YeZEK7tBRFev Xm3rqTuoA+cxMnstznHwXuUa5iCK4ddEvuExgkNEyUWnV6+fHDdunN+u4BuzyTbKNxmhgoVT vgXjVxh36/5/mMLsmmaLAqkVZhlUQlizZs1Cs6r7AmKRWlwKs7QbyHOQozAUYsiQIX4bgk9d yUaFWUhk+V6PuKbiiEt3KNxRWKncQW7aBZOuOHoERsK5JwESyG0CeiS3Y8eOcTCwhkg++YAX F759yI0EgghIhzs/CrPujElHL2yfaGYE9VQ+sxi0zk46X1Da4HW1Tp06tqOVzhIDKWcQg5J8 Ld1y4/M0MmAB2UIGWsbov+hZoyCGlG8Qlcxeo3wzyzMTqaWiMOM5rmk2fl9aYdZLrxA2a9as PNnTaaBdFJ8Duo0uyhlmac+RX/hCcDc4DMTgqLQlekBB6z1heXbTi/p5Ts4wo+G/+uqrfSEn 64DC9EgqhLvGUBRmhIuHUC10/ZJBHBkx0nF4TAIkkNsE9CAe2gk4S8KsMrxSduvWzW9/0OlN NsuX2yRzu/TS4S4shVlmh4MoQylD5ymsjsL7NmZW5F2KPeLDAiu/m5Qzv/dle/yClBvykUEK LYPWrVsnVZbBjfIt/NpD+RY+4/w+IVWFGelq02z8xrTCjPAXXnghrv0bNmxYDI4W169fH/cZ JtyrBx2LS2EeOXKkn1+sWYb5NdY1b9iwwToC0+2Im2cqzJB4CdmGDx/uV4REL+3Zs2f78VAh 3BF2rTA3bdo0zxogPWIUtj6xhCBlMUiABApAQJtsuS8inGNWiN6xCwA4B26VDnd+FOaixoKB IJh6f/nll0mXQ4XlTcoZFl5Sr2ei3PjMF/jDxB5KcKY3yjd9opRv+uwK604ozPJJPD2DGvQ8 d+DbVZgxgda5c+c4nSLoXQ+9QvsTKC6FGd9i1zPIQXnV1/CNaPhbwUaFOaiGZOk11zwCQsdI K9YYQ0nGSJA42ZEKgXXKuhKj6FphRrwmTZrYjgBmhrBgX+7FfsaMGVlKi9kmARIoCgJfeJ6K 9Wd9pP0488wzYwjjRgKJCEiHO8oKc6L8pxom5Uw1fkmJlyvlzpVyuvUyV8qdTeXUM8x6ja4r OznXk2Rh/kZgji1KuLzjsYdPh5kzZ0pS/h7vfokXtqxzypQpfpyggXUdjplidwsLh9l1gwYN /LQlH1CkMQP9+++/xxo3bmzDYV0kXwvS6SWyTHLzEeXzUsicByAnN29Gx3iVNqWyexXBLFiw wOy5555x8du2bWsGDhwYdy3oxDPrNu3atQsK4jUSIAESiCPgvSBNqVKl7B8CDj300LhwnpBA EAFvVN9enrF+ht0/1fypoGhZf03K6XXksr4s+SlArpQ7V8rpyj5Xyp0r5XTl6557Fh7Gm3W2 f6VLlzb777+/GyUy557puPGW1RjkE2pj5cqV/f5JZDJZyBnJaYUZbFEJnn32WeN9Ny0QtTeK Yvr06WMaNWpkypQpkyeOKMyI55lr2HQ+//xzPx6uv/jii8ZbO+Rf4wEJkAAJkAAJZJqAdESp MGeabDTSE/mW9IGCXCmnW6typdy5Uk5XvjzPbgI5rzCL+Dy7e+OZURvP/MLssccexluLYCpV qmT23ntviRK41wozZoUw+uKtETLlypWzaR1yyCFml112CbyXF0mABEiABEggUwSkI5qp9KKe TklXHF3+lK9LpGSdU74lS54sTckiQIW5gPLUCrPnRCOPyXYBk+ftJEACJEACJJASAXa4U8KU tZEo36wVXUoZp3xTwsRIJFAsBKgwFxA7FeYCAuTtJEACJEACJEACJEACJEACJBBRAlSYCygY KswFBMjbSYAESIAESIAESIAESIAESCCiBKgwF1AwLVq0ME8//bSBcy+aZBcQJm8nARIgARIg ARIgARIgARIggQgRoMJcQGFs27bNOvdCMhUrVixgarydBEiABEiABEiABEiABEiABEggKgSo MEdFEswHCZAACZAACZAACZAACZAACZBApAhQYY6UOJgZEiABEiABEiABEiABEiABEiCBqBCg whwVSTAfJEACJEACJEACJEACJEACJEACkSJAhTlS4mBmSIAESIAESIAESIAESIAESIAEokKA CnNUJMF8kAAJkAAJkAAJkAAJkAAJkAAJRIoAFeZIiYOZIQESIAESIAESIAESIAESIAESiAoB KsxRkQTzQQIkQAIkQAIkQAIkQAIkQAIkECkCVJgjJQ5mhgRIgARIgARIgARIgARIgARIICoE qDBHRRLMBwmQAAmQAAmQAAmQAAmQAAmQQKQIUGGOlDiYGRIgARIgARIgARIgARIgARIggagQ KPXTzxtiUckM80ECJEACJEACJEACJEACJEACJEACUSFAhTkqkmA+SIAESIAESIAESIAESIAE SIAEIkWg1OYt2znDHCmRMDMkQAIkQAIkQAIkQAIkQAIkQAJRIECFOQpSYB5IgARIgARIgARI gARIgARIgAQiR4AKc+REwgyRAAmQAAmQAAmQAAmQAAmQAAlEgQAV5ihIgXkgARIgARIgARIg ARIgARIgARKIHAEqzJETCTNEAiRAAiRAAiRAAiRAAiRAAiQQBQJUmKMgBeaBBEiABEiABEiA BEiABEiABEggcgSoMEdOJMwQCZAACZAACZAACZAACZAACZBAFAhQYY6CFJgHEiABEiABEiAB EiABEiABEiCByBGgwhw5kTBDJEACJEACJEACJEACJEACJEACUSBAhTkKUmAeSIAESIAESIAE SIAESIAESIAEIkeACnPkRMIMkQAJkAAJkAAJkAAJkAAJkAAJRIEAFeYoSIF5IAESIAESIAES IAESIAESIAESiBwBKsyREwkzRAIkQAIkQAL5JzB18vj835TFd9S97Moszn3+s0755p9ZNt1B +WaTtJjXXCNAhTnXJM7ykgAJkAAJlEgC7HCXSLH6haJ8fRQl8oDyLZFiZaFKCIGcVZg3bFhv YrGYKVWqlKlUac+k4vxl82azfcd2s+suu5q/VayYNH5xRpCyJcoDylG6TBmzxx57JIrGsAwS 2Lxpk9nxxw5TevfSplz58hlMmUmVVALbtm0z33+/1pT36stee+1dUovJcmWIgHS4S/rMa66U 060WuVLuXCkn5Zt9FiLox636dJX5cd06Kz7oEPsfeKA5/PDDTcWKlVyR+ufS//MvOAfok++2 226mbLlyTshfp3/88YfZtGmjfyE/+ojWC8qULpPwOf4DeBBHICcV5t9+/dVUr3aK+fKLLyyM OfMWmRNOPDEOjHtyW8vm5vnnhpsaNc82r02bYXbddVc3SiTO3bIly1T5ChXM/Q90M01vaWbK eAo0t8IhsHXrVlOrxhnm45UrTd16/zQvj37V7LLLLoXzMKZaYgg88nAf8+D9nW27M3X6m6wz JUayhVOQXFE0cqWcbi3JlXLnSjkp3+xRmH/77Tfz9JNDTJf77nHF5p+3u7ODaXtHO7PPPvv6 13Cg+39xASEnLW9rbVrd1tYc5inhepsw/lXT+LqG/qUj//EPM3/RUqto+xcDDlZ98ok57eQT /JAevfuatre38895kBqBnFSYUXnPO6eGWfbRR5YSKt37Hyw0//d//xdKrcOdd5gnhww2V13d wAx//sXIdlzdsoUWyAnAD3OuN3DAmU8HTIZOMZBxbq2zfIV51CtjIzvokqEiM5kCEpj51pvm 8nqX2FSi3u4UsKi8PUMECqJowOLq1bGjTcc725sff/zB5ujkk08x3Xr0Muedf0FgDhfMn2fu ubuDmTvnfRuODuNdHe4OtMJauGC+mTxpojnrrBrm4rqXBqaX6sWClDPVZ0QxXkHKTflGUaLx eaJ843lE5ey///3R1L3oAtt/SyVP7iRcuv3yKa+/Yc45t7b/SFdhRsDij1aavx95pB8n6GDI oIHm7g7t/aBH+j9umt/a0j/nQWoEqDD/j1PPPg+bNm3vCKUmCnPUZwf1D/Ogg6qYES++lEcx 22333c26tWvNCG/GfPKkCX6ZH+ja3dzVsZN/zoPMEfj999/NP+teZDuWNzS5yQwc/GRkB10y V2qmlC4B/C6vu7aBf3vU2x0/ozwoVgLpdrh37Nhh7mp/uxn2zNOB+e/Vt59p3eb2uDA9oANL JSxbwobB13kLl8QNQG/fvt3UOe8cs3DhAvOfqdPNubXPi0srvyfpljO/z4la/HTLTflGTZLB +aF8g7kU51UMNLVs0cyMfGGEn42+/fqbWueca/baey/z6y+/mlnvvG3uaHubH+5OQOl+OSJh hrdy5cp+fBz8svkX8+mqT8zjj/X3r++7735m0dJl/rLRIIV50JCnzI03NfXvcQ902ythVJiF RP72VJgVL3dUSAWZbFSY6ze41jw74gW7TluXRR+/PGqkada0ib3k/jh1PB6TAAkUPgGsM+r5 UDfzxOBBcQ+jwhyHgychBNLtcI984Xlza/ObbaoPdnvIXNOwkdm6ZasZ0L+fXYqEgGkzZtql ATj+888/7QDgu7PeMVCmb2vd1vz003/9WZjRY8fbpSeIi+21/0w21za4yl7LhHVNuuXcmZvs /Z9uuSnf7JA55Rs9Oa1b97058rCDbcagCM+YOcvsv/8BeTKKeJgUwbI7bOMm/sfUuehie6wV ZkxkLVq63JQtW9aGuf+++vJLzwK2pm/lA4tW9OWxuQozLID28CxjsVwL65+DthXLl5vqp58c F0SFOQ5HyidUmBWqRKbZqSjMWJC/du13Ztvv26ySijWqB1WpElqRMeqLTSo6RujXejO/6IzA PLzyQQf5YSqbCQ/1DzOVTrbu+CRSmFG2b77+2vzqmRbv7s1Ql69Q3hx4YOWEyjgyCgVg7Xdr 7WwqnCPss+8+oc6LXB5r1nxjNm7Y6eCggjeDcfAhhyR9HtaZfPftt6Z06dIGs7rlypfzRvIO CmWGZyJfsiYdpjc///SzjZ9KGfG8b9esMRjFgxwT3QOGGK0UeYdmKp8Bwhhpw4nbPvvsE2gS mc9kGb2ICcx+d5ZVONzH4nd5+hlncN27C4bneQik0+HWMxB3tL/LdPdMsGXD++Ga+leaaVNf My1atjIPPzLAtpeyxOSn//5k5i1a7K/ZE8VMd8j0O+md9+aaU089TZJPe59OOdN+WIRuTKfc lG+EBJgkK5RvEkDFEKzX/yabzdVWN9oqR7eBifrZUrzRr4wyNze5wZ7qttRVmCV+IrPsAY/2 y7PuWqcpaXCfnEDOK8wdO91rsA7rrTdnWFq6kmt8yRTm11+bYlrd2twfFZJ78eMYOPgJU++y y+WS3X/91Vfm+GN2rjt4653ZZtHChdYkLi6SdzL4iadN4xubJFUU5T79w0xFYYaSBScCEyeM M2E/5Pffm21aNGvqO0mTZyF+74f7mQbXNMyTv2+/XWO6d30wzoxF7ruwzkWmR6++5rjjj5dL RvPATMYb06eZfn17++E4OO20081TQ4ebo485Ju46TqAcDx821FoCuIG47zFPBiedFD/KJrMe GDV8dfxk89iAR/zZFEkDZXx+5Chzdq1z5JLdJ3IAgXueeOqZuHV6Wi4YLRz23PNm0sTxvgMH d0ZGHgb5tG7VwubLlQ/y0KdXD/Nov74S3d+jznS6t7M55JBD/Ws8iDaB8ePGmhuub2QzWe2M 6gYv51YtbrFmrKn8lqNdOuauKAik0+HWbS8cyBxz7LFxWZ32+lRT/8rLDWZGFny41BsUrGCd 2MAPyLrv18WZDMKkG6aJumMp9VravUw4O0ynnHGFytKTdMpN+WaPsCnf6MlKK8zugKKbW/R7 293exhxx+BG2/ye+H3T/z+3HuWng/JOPPzann7LTEbFWbkVhRhvc+f4HzSN9+1idQ7e3Oj39 XPQH35v9roFVkE5Tx+dxYgI5rzA/PWy4N+J9ul85gSvINDtMYYZCc1+njmbg4wMSkoYC8/ig J/zZxc9WrzYnnxjfMQlLID+VW/9AUnEUpF+mQT/kZ5560rS/o01Y1ux1eNt7qFcfX2nWJiyJ btSds/zwmP7m2+asGjX9pDHDCocM4sTND3AOhjz5jB18kMvS+Mh5ov3MWe+Z06udYaPoWflE 92hTGpmRES/ZMEuE6c1JJ+xU/sNkpVkiDkzsMRuOTw3VOf/cPIMYOj9oVN+e9X7gAIOOx+No EEB9bH97W0/ZeNJccmk9A4uEGtVP8x3F0bN6NOQU5Vyk0+FesmSxOfvMaqEDpp9/9pltp/B+ kNlkec+s/3m956V1ianwt79ZLNKmyjtLt3tY13zsccdlBF865czIg4s5kXTKTfkWs9Dy8XjK Nx+wiigqrA7POPVkfzIMSjOsbapU2WmmnUo2pL1EHzWon+2mIZM5uC5tKY6lfcUxJtoe8SZL pkyeZM70HCkGmWXLbx99QXzdp12b2+wAvE4TaXFLjUDOK8xScTBL90CX+yw1mGbj5Q7TY9nC FGYxQZN4mI28pG49E/NM2WbMmG6ub3iNBMVVfFdBRIUe8PggU/u88+1s6SBPAZd1jJgBneN5 8UacZJv+YWJWKmi9GDriv3uewpct+8jcfNMNvtLlxv9g7hxzoeeoRTYonFdedbU1r5437wPT 9MbGfiOilcOhTz/ljbK1trdhPdz1jW8we+65l1m//mfz3LPDTI/uXW0YnF9hZAwm0S4PRBg7 bqK56JK6ZsuWLQY8und9wN6HmQ7ppGHAoknj6824V8fYMDRGY8ZNMFW92eRfftlsZ2Y73/uX IzNtEqgbH9yMdJ94eqg56eST7cxJ0yb/9pVwnVc4rql99ln2eVdeVd90eaCrNRff5s1yT5s2 1XKR9NwZGTSYMluIcsvsPmSLMrmNsMzOID1ZE+Mq7JiNhCXCP446yirSGOSQWWfX+YTNNP9F kgB+l5h9Q73A5v6WqTBHUmyRylQ6HW4x/9Ptqi6Ubpul/dSKsB74vL/zvab/Iw/777oXRjzn WV41M7r9xAAnrGOwpbKsR+dFjtMpp9ybzft0yk35Zo/EKd9oyko+76hzd/4FF9q+/unVqlkv 1XvttbcOjjvW7/KwdlZucD1yY1Kv0XX/tsG6zwodZenSxeaWm260YR8uXWGgu+itd8+HbH+7 hvc53PHemuoLzqtl+7Si9+i4PE5OgArz/9yrowMglQnYXNPsIIUZL/5aNar7Cues9z8wp5xy ahz15cuWmTO9bz5jgzI3Z/5C6zBAd0IQ5t6rlSI9so+4iTb9w0wUzw2DwgalHAoWNiiija6t b0evcI7RLChmesPs51nVTrNKswwyYH2umBAjrYWLl9n1xHIfynW+59AASicaDlEoXR5BnlQf 6vagNUFGWqKgywgaruF578yek2eNNJTpG/99HaJYZfWll8fYmX7d+CAvcxcs8r0RIi6c2FQ7 5SRbPnwGBQoLyicdEMRZuvwTc/gRR+DQ3zDwIgqrWCtouYjCDOVIzB1xs24Yca7rgM6fVtgx 6wzzbnddtB4ACjP3xjO4RZdAWJ2Jbo6Zs+ImkE6HW9rBMCsXXQ/x+cUTq1a1xZR3Ipa8PPLY QPPxihW+4zAo0fACW/Osavb9+NGKVabKwQfbQU9pG5EI2rVJU143Rx19dL7QpVPOfD0gopHT KTflG1FhBmSL8g2AEoFLO3bsML16dDd9e/cMzQ0ccN3aqrVnil3X9+kgkXUbimuwkjzggANt P1viwD8Q/Jh0vOuv7yO7s9HyW8Y9aIsrVqzoL+10lWA9qIm+5TXXNrK6CKwc3biSB+4TE6DC rL5HphURYBNlB8fSOdDKjl7bcGur26xDFMR1N3i9xY8Nm4zQawUxbG2X/nHIfW7a7rn7w3TD g85hLt61ew+r0Eu4zh86Us+9MNKf+ZI42GvlURwP3Nayub8WuM/Dj5qr6te3jYPcB3Ni6zis fAVTpkwZe1k/D4yDZsaxPuSYI3cq9BIH3w0V5wgvvTLGXHb5v+Qx/l47PdENkOaLGVrMgrgb zNExY6vv02WG3Ju3aGVH9mRmEM7bNm3eZBV3OOHCpuWi69CmjRv9TmUNbxQQ390T5VdMIXE/ 1p/c12XnDLvURVzXHVicy4bn1apxhjXn1c56JJz76BMIqzPRzzlzWFwECrvDrWeT3ZkQKbMM Nsu3P29vd6d5qGdv6/0d70EMzna8+x7rpwLr6aA0y8CppJFsn045k6WZDeHplFvec6kMiFC+ xVsLKN/i5Z/s6VA2R774vIEjrUSb2xfV7/JE97lhEya/Zi64sI5/WX7LuACdAEr6xReeZz9X 6ppliz6D9nbuvEVm/wMO8Lxv1+AMs08z/wdUmJXCDHxauZVZUyh3oqRoZUdXXr3G1RXD0qVL TM3qp9vLMjOqFcSwBfs6/XQVZihLu+2609386s9WW2+nOn+i5OprONbKGpTFXn0fNlt+2xIX DR6oMVsga4clj3r9hdyAH3bD6/9tzjmntl3HJkqhhGsewkjCZI9G5xKvcUBDIC9/fIz9ySGD 4xRaia/32nReOgWa77tz5tnGR9+DY5G7Vpix9viEY+NNX9AoNW/R0lzsmZBXrXpSHvN53WDq OoRniBkjjrVZjTZt14qx5Anx8e3wE7wZH3hmlw2K+2bPHP2ejnfZSzK4IJ7AJR730SaQqM5E O+fMXXERSKfDLcs+3AE7KYMeqHQtjTDzAkcyP/74gx1QPemkU+zgoVjnbNm6xS412Wefff0B PHlX6vZclpvIM5Pt0ylnsjSzITydclO+2SDZnXmkfLNDVlg+ha/GfPLxSrsMD5Mq7jby5dHm 8iuutJf1u9yNF3SOSawOHe/JY72o+6zS35ZlL0hH9x+lXyn9P0wcUWEOop36NSrMjsIcZpot SopWdoIqbxB6rQyKKUTQNffeVNPX9+kfJmau3e8wQ9mrf9Xl/rfiwsyYtXt8nX6iY/kBw5wb Aw9YPxG2YcYUzsLEWYzmoZVD936RA/INE/KuD3ZJSWEOYhl0Lex5WmFGnP9MnmgaXVPfje6f Q6HH+m0x19Zy0XUIN2iz8n6PPmYdSsDr98UX1LaDA3rgBo3epRdfaEcU/YclOZAGkwpzElAR C05UZyKWVWYnIgTS6XDL58zcNk6KJIOnMlNx6GGHSVDoXpaEiGWMWGNhNln8T+DmQQMfswN7 MisdmqATkE45nSSy8jSdclO+2SNqyjd7ZKVzCp8M8z6YaydY5DvM6KNinTE+Eavf5bgPyjS+ 5Yy+smwwyYaJ9RF//3ueZYUSJ6jP+sXnn5uqx+9c0iL6hTbHhg6AL9noPEg8SZf71AhQYXYU ZmATUwZBuGjJcvPiCyPsbKpWdoIqr9yj99LhwDWpqFpBlGv6Hhynmr6+T/8owhQljP7XrlXD X3tdwzEFRnpaIURHqt5ll5lNGzfpR8UdYybh4X79je5MoYwow+CBj/vOwfRNUARh9g6zZc0j TGFG43Jnu7bWRBrrqeH1795OHfKYTOtnyHEQy6BrEl/2oqAHdSZhkjh1yhQz/NmhZr7nBM3d 0MGUdeGuXLQDJ20yLsrxypUrfKsEbYGg08HzrvjXVf43pN3ny3n1M88yrVon9nQucbmPDgEt a93uRCeHzEnUCKTT4Za2F+1VkONBcf4Y1AYGlf+77741R//9sECfHW49HjP6ZeskUc/GBKXp XkunnG4a2XieTrkp3+yRNOUbLVnBguaVUS8ZtGmlS5c2be9oH7g0UXKtJ9zQns5buNgcfPAh ccoqBg0XLV1uypYtK7elvA/qs2pfN2KWvXDBfOuwV7fpuj8RpnOknJEcjUiFOUBhRl3Qptkw J97DGyWaO+d938MxHDbpyuuaqun6JB0OXJOOgbzEcC2s8ur0ZfYW8RNt+kfhdk70fXpWE9fd PGgl33VGpdNJ9Xjt2u/MSs8pDL49jG91yibl0jzCzPNQNlmXG2SSLZ88kbT1Hh7H4UwBDcjs 9+dZr4ap8E2kMOv04QDu01Wr7Lo8Wa+OcOGaTC7aZPyDBYvNdM/bdpf77rH5dTuxkic0vB9+ tMKOYOq88LhkEEhWZ0pGKVmKTBJIp8OtZyOeGvqsue76xnFZck37klmqSHw9a6xnmPV65cGD HjedOtyZx8lmXAYCTtIpZ0AyWXcpnXJTvtkjZso3WrLCb6e657T3yy++sE5lZcY4US71UjrR C/S7PNWBx6BnhPVZdf8RjmhHvfSi9ZukJ810HqRfGvQMXgsnQIU5RGHWI0Uan1ZCpROA8CZN bzYDB+ddx4AwUXBwLLOnWkEMq7xhPw6kE7bpH4XOa1B8MZuTML2eWedP/+gkruyxfuKVl18y xx1/vLm1ZWtTsVJF+zmsxYs/NDW8byW3v6ujRPX3iC+u8MXZln6e/gSJf5N3oHmLo7SxY17x nX7BMdnV9f/6jJfcq5mgsVqybKU1BU+Fr8hOGrnynqMyrFdZMH+e2bhpo+85W56FvfaMLvnc tm2bv34kSC4yK4P7sX4FznDQSMv9GKCRTRyR4VwaZAmTPRR4fKf5jOpn2u+MN72lWcKRUbmP ++gQ0PU2qM5EJ6fMSVQIpNPhRt5FycWA4mTPa7V8c37ypAnmumsb2OKlMnAq5oHSXlaqtKe9 d/OmTabGmafbNk2caaJ+i0+KsEFSe3PAv3TLGZBUVl1Kt9yUb3aImfKNlpxg1ShffUHOZMIr LJeI3+KWplZhDZthdtvGsLSCrof1Wb/+6ivfWzaWOk6f/rpddvniqFesFSLS0v2JMJ0j6Jm8 9hcBKswhCjMQuabZuKY7ru5npabNmGlqeObNetNpYEZQPl2kFcSwyhv249Dpu8f6R6Hz6sbD uR55xjniyyeXYIpS96IL/LWyI158yTraQjzZ9GelpHHY2/sWnYzIobx6vZrcp8slP2jNA/GC PrN1a7ObbUOEcPnslOaLhujdOR9Yr6uIIxtmtO9oe5s91QqozofMdMs9sncV5ooVK5mWLZqZ kZ6JPjYZAJH42OvZefFQjTXJ4nAhSC5oaHW6kl5QR1LWpCGOrOUGf73J2kBcEy+1OpzH0SeQ n99y9EvDHBYFgXQ73GjLLzzvHH+ZDtoTePuXDaZ+k1+bZpfPyDV3r5fM6GUkEk8rbd2697TL ft56c4Ztr4PeE3Jf0D7dcgallU3X0i035ZsdUqZ8oycnvTwRuUNf+F9XXm30JAauo8/87NBn 7LJBnKNvhuWccBqs3+WFoTCLWTacL8oaajxHPmOL/Og8hOkciMctnAAV5gQKM7Bp02ycu8rO y6NGmmZNmyDIbkOefMZc+s9/2k8DwXFW4+saSpDRnQitIIZV3lQUOj/x/x3oH4WbVzcuzmFm jplI2bSHaq2YIRxOrK5vfIMpV7ac98H0Jfabm5gFxaaVMukY4bp8o/OYo48x27ZvM2/PfMvc cH0jBFlz48WeSTGcH2geNtD7N3bcRFOtenW7drrrA10MZpOxYZ0vPF3D07araKKRQIN2wokn 2gZs+LCh9tuf9kbvn1bEU+HrKsyYMdHfTsbzHh88xNQ651yzS6ldLBco59JoiWKfilxcR2sY cNDmi1IGdzADeRg1eqw54YQT7eesRgx/1n6sXuIvW/lp3Npyuc59tAmkUmeiXQLmrqgJpNvh Rj6hVDW/+SYDJVZvGPR7qGefhMqy3H9WtdMMvp4A3w3uIB4c4+CrBs89O8xPHm3X1OlvmqOP Oca/lspBQcqZSvpRjVOQclO+UZXqX/mifP9iEZUjUUZh9Scb2ra7O93r9TOrWsUZ/b1+fXvH +eoZ8+oEc8ml9ewt+l2ONm/R0mVGrG8kzVT2ifqs2iwbaenJIZzrPITpHIjHLZxAzirMshY2 WcUJmoXVDpugsD3U7cGEHzQHfnyvt3ffR3wHTVpBDMtDoh9HmEjxo5CyQWEO+p6xe68ohbiO H/Os9+eaKlUOttFgfgwT4EQbFLv3Pphv9t57HxsNzrDOO6emP1sRdi++/Swm25pHWHxcR/7c WeSNGzeYK+rVtRYBie51TbZT4Svmz7qRg8zF7CbR82ooZ2q6HoXJRcdBuuJhNugZ+NTLJXXO T8pYTN6D0uC1aBPQ9SGszkS7BMxdURMoSIdb8rpmzTf2M3X4dMrfKv7NDmhKWCb233+/1rNu +s3stvtu9j3jztSk8oxMlDOV50QtTibKTflGTap/5Yfy/YtFlI7Qx2x6Y2PPt8zrKWULE2dY WiebfpfrvqSEp7pP1GfVZtlIb/TY8XaCT9LWukGYziFxuQ8mkLMKs5jHpqJQaLNfd9RGsL7/ 3mxze5tW/syiXMcMa/eeve0MpFzDXn/bMsjRCuLomUw4Gzj2uONwOeGmR5GwFhjrqpN1SFyv 2dpZCx6G9botW9ySp2wI6/Pwo3b9tuvxD+bqcLSFGXp3A5MevfuammfX8oO0wnz3PffZ2eMe 3bv64TjAoEOX+7t6nbiKcddxgllXrI2G2ba7wUFY5/sfNP846qi4IDR+V//rMntN1tXFRfBO ZLbcNX22z/O8J8IxF8xg9IbRx/6PDbSu/MVBjv5MVCK5DHi0n00T6SWTOWZsnn5yiB9f5+HC OheZLg9289Yvn6Yv8ziLCOC3LGs8E9WZLCoSs1rIBDLR4S7kLGYk+VwppwsrV8qdK+WkfHd+ p9jlENXzd96eaSfIYJkZtMEap3WbO6w5tg7X/T/0Jed6X4cpV768jpLSsdYJ3D4rJnIaXVvf TJk8yU4szf9wiT+JhcS1bpCK3pNShnIsUk4qzIUpY4yew4QDf3BDv99++xfm44o0bZkZKF+h vNm6ZavZ/4ADTJkyZRLmAUodZpxhPg0lE/Fhgu1uWmEW03Xch4YGMx3lvcZlL299dLINcTEY gU9VoYHAd+2w7riwNjzvhx/Wme3btpvdS+9u8ImoypUPsuUtrGe66YLxuu+/N2XLlbUzN3C8 lgorNx2ekwAJZDeBXFE0cqWcbm3MlXLnSjkp3+xSmEVemBDa5DkyxGbXKHv94f322y8tJVjS 5D76BKgwR19GOZFDrTDTXCQnRM5CkgAJZJhArigauVJOt3rkSrlzpZyUb3YqzK7ceJ4bBKgw 54acI19KKsyRFxEzSAIkEHEComhEPJsZy17dy3Krw035ZqzqRDIhyjeSYmGmSMASoMLMihAJ AlSYIyEGZoIESCCLCbDDncXCSyHrlG8KkLI4CuWbxcJj1ks8ASrMJV7E2VFAKszZISfmkgRI gARIgARIgARIgARyiQAV5lySdsTLutlzogBPf2XLlStSp1kRx8LskQAJkAAJkAAJkAAJkAAJ FBMBKszFBJ6PJQESIAESIAESIAESIAESIAESiDYBKszRlg9zRwIkQAIkQAIkQAIkQAIkQAIk UEwEqDAXE3g+lgRIgARIgARIgARIgARIgARIINoEqDBHWz7MHQmQAAmQAAmQAAmQAAmQAAmQ QDERoMJcTOD5WBIgARIgARIgARIgARIgARIggWgToMIcbfkwdyRAAiRAAiRAAiRAAiRAAiRA AsVEgApzMYHnY0mABEiABEiABEiABEiABEiABKJNgApztOXD3JEACZAACZAACZAACZAACZAA CRQTASrMxQSejyUBEiABEiABEiABEiABEiABEog2ASrM0ZYPc0cCJEACJEACJEACJEACJEAC JFBMBP4fx/BRetGKoRkAAAAASUVORK5CYII= --------------E141655F6BC3F6F8ABEBA096-- --------------5323FA587AD2C243D1E46CA3--
On Mon, Apr 17, 2017 at 9:12 AM, FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Should I understand then that fecing is not mandatory, just advised and there are no downtimes to the rest of the cluster if a pending manual action has to be done to confirm a host has been rebooted in order to move the SPM to another live host. Perhaps only effect is not be able to add new hosts and connect to storage or something ?
A missing SPM will not affect existing VMs. Running VMs (on hosts other than the SPM host) will continue to run normally and you can even start existing VMs. Without an SPM you will not be able to manipulate the storage (Add/remove domains, add/remove disks, migrate disks, etc).
Fernando On 17/04/2017 10:06, Nir Soffer wrote:
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this [image: Screen Shot 2017-04-16 at 10.22.00 PM.png]
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you configure power management for all hosts? did you check that it work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot move the spm to another host, unless you manually confirm that the SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4, Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
> Fence agent under each node? >
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
> > On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote: > >> On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < >> konrasko@gmail.com> wrote: >> >>> "Corner cases"? >>> I tried to simulate crash of SPM server and ovirt kept trying to >>> reistablished connection to the failed node. >>> >> >> Did you configure fencing? >> >> >>> >>> >>> On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> >>> wrote: >>> >>>> On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> >>>> wrote: >>>> >>>>> On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> >>>>> wrote: >>>>> >>>>>> >>>>>> >>>>>> On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> >>>>>> wrote: >>>>>> >>>>>> On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> >>>>>> wrote: >>>>>> >>>>>>> When you set up a storage domain, you need to specify a host >>>>>>> to perform the initial storage operations, but once the SD is defined, it's >>>>>>> details are in the engine database, and all the hosts get connected to it >>>>>>> directly. If the first host you used to define the SD goes down, all other >>>>>>> hosts will still remain connected and work. SPM is an HA service, and if >>>>>>> the current SPM host goes down, SPM gets started on another host in the DC. >>>>>>> In short, unless your actual NFS exporting host goes down, there is no >>>>>>> outage. >>>>>>> >>>>>> >>>>>> There is no storage outage, but if you shutdown the spm host, >>>>>> the spm host >>>>>> will not move to a new host until the spm host is online again, >>>>>> or you confirm >>>>>> manually that the spm host was rebooted. >>>>>> >>>>>> >>>>>> In a properly configured setup the SBA should take care of >>>>>> that. That's the whole point of HA services >>>>>> >>>>> >>>>> In some cases like power loss or hardware failure, there is no >>>>> way to start >>>>> the spm host, and the system cannot recover automatically. >>>>> >>>> >>>> There are always corner cases, no doubt. But in a normal >>>> situation. where an SPM host goes down because of a hardware failure, it >>>> gets fenced, other hosts contend for SPM and start it. No surprises there. >>>> >>>> >>>>> >>>>> Nir >>>>> >>>>> >>>>>> >>>>>> >>>>>> Nir >>>>>> >>>>>> >>>>>>> >>>>>>> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >>>>>>> konrasko@gmail.com> wrote: >>>>>>> >>>>>>>> Hi Fernando, >>>>>>>> I see each host has direct connection nfs mount, but yes, if >>>>>>>> main host to which I connected nfs storage going down the storage becomes >>>>>>>> unavailable and all vms are down >>>>>>>> >>>>>>>> >>>>>>>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>>>>>>> fernando.frediani@upx.com> wrote: >>>>>>>> >>>>>>>>> Hello Konstantin. >>>>>>>>> >>>>>>>>> That doesn`t make much sense make a whole cluster depend on >>>>>>>>> a single host. From what I know any host talk directly to NFS Storage Array >>>>>>>>> or whatever other Shared Storage you have. >>>>>>>>> Have you tested that host going down if that affects the >>>>>>>>> other with the NFS mounted directlly in a NFS Storage array ? >>>>>>>>> >>>>>>>>> Fernando >>>>>>>>> >>>>>>>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>>> konrasko@gmail.com>: >>>>>>>>> >>>>>>>>>> In ovirt you have to attach storage through specific host. >>>>>>>>>> If host goes down storage is not available. >>>>>>>>>> >>>>>>>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>>>>>>> fernando.frediani@upx.com> wrote: >>>>>>>>>> >>>>>>>>>>> Well, make it not go through host1 and dedicate a storage >>>>>>>>>>> server for running NFS and make both hosts connect to it. >>>>>>>>>>> In my view NFS is much easier to manage than any other >>>>>>>>>>> type of storage, specially FC and iSCSI and performance is pretty much the >>>>>>>>>>> same, so you won`t get better results other than management going to other >>>>>>>>>>> type. >>>>>>>>>>> >>>>>>>>>>> Fernando >>>>>>>>>>> >>>>>>>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>>>>> konrasko@gmail.com>: >>>>>>>>>>> >>>>>>>>>>>> Hi guys, >>>>>>>>>>>> I have one nfs storage, >>>>>>>>>>>> it's connected through host1. >>>>>>>>>>>> host2 also has access to it, I can easily migrate >>>>>>>>>>>> vms between them. >>>>>>>>>>>> >>>>>>>>>>>> The question is - if host1 is down - all infrastructure >>>>>>>>>>>> is down, since all traffic goes through host1, >>>>>>>>>>>> is there any way in oVirt to use redundant storage? >>>>>>>>>>>> >>>>>>>>>>>> Only glusterfs? >>>>>>>>>>>> >>>>>>>>>>>> Thanks >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> _______________________________________________ >>>>>>>>>>>> Users mailing list >>>>>>>>>>>> Users@ovirt.org >>>>>>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Users mailing list >>>>>>>> Users@ovirt.org >>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>> >>>>>>>> >>>>>>> _______________________________________________ >>>>>>> Users mailing list >>>>>>> Users@ovirt.org >>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Users mailing list >>>>>> Users@ovirt.org >>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>> >>>>>> >>>>>>
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
-- Adam Litke
This is a multi-part message in MIME format. --------------52C86E47EED6E2100703EC63 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Thanks for the clarification Adam. That makes total sense. Reagrds Fernando On 17/04/2017 11:28, Adam Litke wrote:
On Mon, Apr 17, 2017 at 9:12 AM, FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Should I understand then that fecing is not mandatory, just advised and there are no downtimes to the rest of the cluster if a pending manual action has to be done to confirm a host has been rebooted in order to move the SPM to another live host. Perhaps only effect is not be able to add new hosts and connect to storage or something ?
A missing SPM will not affect existing VMs. Running VMs (on hosts other than the SPM host) will continue to run normally and you can even start existing VMs. Without an SPM you will not be able to manipulate the storage (Add/remove domains, add/remove disks, migrate disks, etc).
Fernando
On 17/04/2017 10:06, Nir Soffer wrote:
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
But actually, it didn't work well. After main SPM host went down I see this Screen Shot 2017-04-16 at 10.22.00 PM.png
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you configure power management for all hosts? did you check that it work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot move the spm to another host, unless you manually confirm that the SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Oh, fence agent works fine if I select ilo4, Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
"Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com <mailto:nsoffer@redhat.com>> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com <mailto:dyasny@redhat.com>> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com <mailto:nsoffer@redhat.com>> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com <mailto:dyasny@gmail.com>> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>>:
In ovirt you have to attach storage through specific host.
If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>>:
Hi guys,
I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users <http://lists.ovirt.org/mailman/listinfo/users>
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users <http://lists.ovirt.org/mailman/listinfo/users>
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users <http://lists.ovirt.org/mailman/listinfo/users>
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users <http://lists.ovirt.org/mailman/listinfo/users>
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users <http://lists.ovirt.org/mailman/listinfo/users>
-- Adam Litke
--------------52C86E47EED6E2100703EC63 Content-Type: multipart/related; boundary="------------627018FE70FE823EEDB7FEB5" --------------627018FE70FE823EEDB7FEB5 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit <html> <head> <meta content="text/html; charset=utf-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <p>Thanks for the clarification Adam.</p> <p>That makes total sense.</p> <p>Reagrds<br> Fernando<br> </p> <div class="moz-cite-prefix">On 17/04/2017 11:28, Adam Litke wrote:<br> </div> <blockquote cite="mid:CAG8F9UE5HVYeY0iwQ+1cJmaMHDx0nLnZywrJayGq8WBdTDV+Ew@mail.gmail.com" type="cite"> <div dir="ltr"><br> <div class="gmail_extra"><br> <div class="gmail_quote">On Mon, Apr 17, 2017 at 9:12 AM, FERNANDO FREDIANI <span dir="ltr"><<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com" target="_blank">fernando.frediani@upx.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div bgcolor="#FFFFFF" text="#000000"> <p>Should I understand then that fecing is not mandatory, just advised and there are no downtimes to the rest of the cluster if a pending manual action has to be done to confirm a host has been rebooted in order to move the SPM to another live host. Perhaps only effect is not be able to add new hosts and connect to storage or something ?</p> </div> </blockquote> <div>A missing SPM will not affect existing VMs. Running VMs (on hosts other than the SPM host) will continue to run normally and you can even start existing VMs. Without an SPM you will not be able to manipulate the storage (Add/remove domains, add/remove disks, migrate disks, etc).<br> </div> <div><br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div bgcolor="#FFFFFF" text="#000000"><span class="HOEnZb"><font color="#888888"> <p>Fernando<br> </p> </font></span> <div> <div class="h5"> <div class="m_7118500493255933245moz-cite-prefix">On 17/04/2017 10:06, Nir Soffer wrote:<br> </div> <blockquote type="cite"> <div dir="ltr"> <div class="gmail_quote"> <div dir="ltr">On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div dir="ltr">But actually, it didn't work well. After main SPM host went down I see this <div><img src="cid:part3.327B103E.0FE862E1@upx.com" alt="Screen Shot 2017-04-16 at 10.22.00 PM.png" style="max-width:100%"></div> <div><br> 2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.<wbr>vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-<wbr>03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1'<br> </div> <div>2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.<wbr>vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-<wbr>03a970dbbec1] SPM selection - vds seems as spm 'tank5'</div> <div>2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.<wbr>vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-<wbr>03a970dbbec1] spm vds is non responsive, stopping spm selection.</div> <div><br> </div> <div>So that means only if BMC is up it's possible to automatically switch SPM host?</div> </div> </blockquote> <div><br> </div> <div>BMC?</div> <div><br> </div> <div>If your SPM is no responsive, the system will try to fence it. Did you </div> <div>configure power management for all hosts? did you check that it </div> <div>work? How did you simulate non-responsive host? </div> <div><br> </div> <div>If power management is not configured or fail, the system cannot</div> <div>move the spm to another host, unless you manually confirm that the</div> <div>SPM host was rebooted.</div> <div><br> </div> <div>Nir</div> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div dir="ltr"> <div><br> </div> <div>Thanks</div> </div> <div class="gmail_extra"><br> <div class="gmail_quote">On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <span dir="ltr"><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Oh, fence agent works fine if I select ilo4,</div> <div>Thank you for your help!</div> <div class="m_7118500493255933245m_-7049131619673457077HOEnZb"> <div class="m_7118500493255933245m_-7049131619673457077h5"> <div><br> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Makes sense. </div> <div>I was trying to set it up, but doesn't work with our staging hardware. </div> <div>We have old ilo100, I'll try again. </div> <div>Thanks! <div> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886h5"><br> </div> </div> </div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual</div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886h5"> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Fence agent under each node?</div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?</div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743HOEnZb"> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743h5"> <div><br> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>"Corner cases"? </div> <div>I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node. </div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>Did you configure fencing?</div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210HOEnZb"> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210h5"> <div><br> </div> <div><br> <div class="gmail_quote"> <div>On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_extra"> <div class="gmail_quote">On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <span><<a moz-do-not-send="true" href="mailto:nsoffer@redhat.com" target="_blank">nsoffer@redhat.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"><span> <div>On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@redhat.com" target="_blank">dyasny@redhat.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div><br> <div class="gmail_extra"><br> <div class="gmail_quote">On Apr 16, 2017 7:01 AM, "Nir Soffer" <<a moz-do-not-send="true" href="mailto:nsoffer@redhat.com" target="_blank">nsoffer@redhat.com</a>> wrote:<br type="attribution"> <blockquote class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354quoted-text"> <div>On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <<a moz-do-not-send="true" href="mailto:dyasny@gmail.com" target="_blank">dyasny@gmail.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage. </div> </blockquote> <div><br> </div> </div> <div>There is no storage outage, but if you shutdown the spm host, the spm host</div> <div>will not move to a new host until the spm host is online again, or you confirm</div> <div>manually that the spm host was rebooted.</div> </div> </div> </blockquote> </div> </div> </div> <div><br> </div> </div> <div> <div>In a properly configured setup the SBA should take care of that. That's the whole point of HA services </div> </div> </blockquote> <div><br> </div> </span> <div>In some cases like power loss or hardware failure, there is no way to start</div> <div>the spm host, and the system cannot recover automatically.</div> </div> </div> </blockquote> <div><br> </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div>There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there. </div> </div> </div> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"><span class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191HOEnZb"><font color="#888888"> <div><br> </div> <div>Nir</div> </font></span> <div> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191h5"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div><br> </div> <div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div class="gmail_quote"><font color="#888888"> <div><br> </div> <div>Nir</div> </font> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354elided-text"> <div> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div class="gmail_extra"><br> <div class="gmail_quote">On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Hi Fernando,</div> <div>I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down</div> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354m_6867735599077560094m_2305649857973554660HOEnZb"> <div class="m_7118500493255933245m_-7049131619673457077m_-3447473577979970513m_1774108031932592886m_3846525510791616528m_1120357608056684743m_362375030716496662m_-7954130722235415210m_2173925241577336935m_-6434846878319371191m_5622908635907641408m_8346864507156293354m_6867735599077560094m_2305649857973554660h5"> <div><br> </div> <div><br> <div class="gmail_quote"> <div>On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com" target="_blank">fernando.frediani@upx.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div> <div>Hello Konstantin.<br> <br> </div> That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.<br> </div> Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?<br> <br> </div> </div> <div>Fernando<br> </div> <div class="gmail_extra"><br> <div class="gmail_quote">2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span>:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>In ovirt you have to attach storage through specific host. </div> <div>If host goes down storage is not available. </div> <div><br> <div class="gmail_quote"> <div>On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com" target="_blank">fernando.frediani@upx.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div>Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.<br> </div> In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.<br> <br> </div> Fernando<br> </div> <div class="gmail_extra"><br> </div> <div class="gmail_extra"> <div class="gmail_quote">2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span>:<br> </div> </div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Hi guys, <div>I have one nfs storage,</div> <div>it's connected through host1.</div> <div>host2 also has access to it, I can easily migrate vms between them.</div> <div><br> </div> <div>The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,</div> <div>is there any way in oVirt to use redundant storage?</div> <div><br> </div> <div>Only glusterfs?</div> <div><br> </div> <div>Thanks</div> <div><br> </div> </div> <br> </blockquote> </div> </div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">______________________________<wbr>_________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br> <br> </blockquote> </div> </div> </blockquote> </div> </div> </blockquote> </div> <br> </div> </blockquote> </div> </div> </div> </div> <br> ______________________________<wbr>_________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br> <br> </blockquote> </div> <br> </div> ______________________________<wbr>_________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br> </blockquote> </div> </div> </div> <br> ______________________________<wbr>_________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br> <br> </blockquote> </div> <br> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> <br> </div> </blockquote> </div> </div> </blockquote> <br> </div> </div> </div> <br> ______________________________<wbr>_________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/<wbr>mailman/listinfo/users</a><br> <br> </blockquote> </div> <br> <br clear="all"> <br> -- <br> <div class="gmail_signature" data-smartmail="gmail_signature"> <div dir="ltr">Adam Litke</div> </div> </div> </div> </blockquote> <br> </body> </html> --------------627018FE70FE823EEDB7FEB5 Content-Type: image/png Content-Transfer-Encoding: base64 Content-ID: <part3.327B103E.0FE862E1@upx.com> iVBORw0KGgoAAAANSUhEUgAAA8wAAACoCAYAAAA1pWe/AAAME2lDQ1BJQ0MgUHJvZmlsZQAA SImVVwdYU8kWnltSCAktEAEpoTdBepXeO9LBRkgChBIgIajYkUUF14KKBURFV0RUXAsga0HE ziLY+wMVlZV1sWBD5U0K6Pra9873zZ3/njnnzH/mnjvfDACK1qy8vGxUCYAcfoEgOtCHmZiU zCT1ARyQAAowoMliC/O8o6LCAJSx/u/y7iZAxP01S3Gsfx3/r6LM4QrZACBREKdyhOwciI8A gKuz8wQFABA6od5gdkGeGA9BrCqABAEg4mKcLsXqYpwqxZMkNrHRvhB7AUCmsliCdAAUxLyZ hex0GEdBzNGaz+HxIa6G2IOdweJAfB/iSTk5uRArkiE2Tf0uTvrfYqaOx2Sx0sexNBeJkP14 wrxs1tz/czn+t+Rki8bm0IeNmiEIihbnDNdtT1ZuqBhTIT7OT42IhFgF4gs8jsRejO9miILi ZPaDbKEvXDPAAPBTc1h+oRBrQcwQZcV5y7AtSyDxhfZoBK8gOFaGUwW50bL4aCE/OyJMFmd5 Bjd4DNdwhf4xYzZpvIBgiGGloUeKMmITpDzRjkJefATEChB3C7NiQmW+D4syfCPGbASiaDFn Q4jfpgkCoqU2mHqOcCwvzIrNkswFawHzKsiIDZL6YolcYWLYGAcO189fygHjcPlxMm4YrC6f aJlvaV52lMweq+FmB0ZL1xk7KCyMGfO9WgALTLoO2KNMVkiUbK53eQVRsVJuOArCgC/wA0wg gi0V5IJMwOsabB6Eb9KRAMACApAOuMBSphnzSJCM8OEzBhSBPyHiAuG4n49klAsKof7LuFb6 tARpktFCiUcWeApxDq6Je+BueBh8esFmizvjLmN+TMWxWYn+RD9iEDGAaDbOgw1ZZ8MmALx/ owuFPRdmJ+bCH8vhWzzCU0IP4RHhBqGXcAfEgyeSKDKrWbxiwQ/MmSAc9MJoAbLsUr/PDjeG rB1wH9wd8ofccQauCSxxe5iJN+4Jc3OA2u8Zisa5fVvLH+cTs/4+H5lewVzBQcYidfzL+I5b /RjF97s14sA+9EdLbDl2GDuPncYuYsexZsDETmEtWCd2QozHK+GJpBLGZouWcMuCcXhjNtYN 1gPWn3+YmyWbX7xewgLunALxz+CbmzdXwEvPKGB6w92Yywzms60mMW2tbZwAEO/t0q3jDUOy ZyOMS990+W0AuJRBZfo3HcsAgGNPAaC/+6YzeA3LfQ0AJ7rZIkGhVCfejgEBUIAi/Cs0gA4w AKYwH1vgCNyAF/AHISASxIIkMBOueAbIgZxng/lgCSgF5WAN2AC2gG1gJ9gD9oNDoBkcB6fB OXAZdIMb4B6si37wAgyBd2AEQRASQkPoiAaiixghFogt4ox4IP5IGBKNJCEpSDrCR0TIfGQp Uo5UIFuQHUg98ityDDmNXER6kDtIHzKAvEY+oRhKRVVRbdQYnYw6o95oKBqLzkDT0Xy0CC1B V6Gb0Fp0H9qEnkYvozfQXvQFOowBTB5jYHqYJeaM+WKRWDKWhgmwhVgZVonVYgewVvidr2G9 2CD2ESfidJyJW8LaDMLjcDaejy/EV+Jb8D14E96BX8P78CH8K4FG0CJYEFwJwYREQjphNqGU UEnYTThKOAv/m37COyKRyCCaEJ3gf5lEzCTOI64kbiU2EtuIPcTHxGESiaRBsiC5kyJJLFIB qZS0mbSPdIp0ldRP+kCWJ+uSbckB5GQyn1xMriTvJZ8kXyU/I4/IKckZybnKRcpx5ObKrZbb Jdcqd0WuX26EokwxobhTYimZlCWUTZQDlLOU+5Q38vLy+vIu8lPlefKL5TfJH5S/IN8n/5Gq QjWn+lKnU0XUVdQ6ahv1DvUNjUYzpnnRkmkFtFW0etoZ2kPaBwW6gpVCsAJHYZFClUKTwlWF l4pyikaK3oozFYsUKxUPK15RHFSSUzJW8lViKS1UqlI6pnRLaViZrmyjHKmco7xSea/yReXn KiQVYxV/FY5KicpOlTMqj+kY3YDuS2fTl9J30c/S+1WJqiaqwaqZquWq+1W7VIfUVNTs1eLV 5qhVqZ1Q62VgDGNGMCObsZpxiHGT8WmC9gTvCdwJKyYcmHB1wnv1iepe6lz1MvVG9RvqnzSY Gv4aWRprNZo1HmjimuaaUzVna9ZontUcnKg60W0ie2LZxEMT72qhWuZa0VrztHZqdWoNa+to B2rnaW/WPqM9qMPQ8dLJ1Fmvc1JnQJeu66HL012ve0r3D6Ya05uZzdzE7GAO6WnpBemJ9Hbo demN6Jvox+kX6zfqPzCgGDgbpBmsN2g3GDLUNQw3nG/YYHjXSM7I2SjDaKPReaP3xibGCcbL jJuNn5uomwSbFJk0mNw3pZl6muab1ppeNyOaOZtlmW016zZHzR3MM8yrzK9YoBaOFjyLrRY9 kwiTXCbxJ9VOumVJtfS2LLRssOyzYliFWRVbNVu9nGw4OXny2snnJ3+1drDOtt5lfc9GxSbE ptim1ea1rbkt27bK9rodzS7AbpFdi90rewt7rn2N/W0HukO4wzKHdocvjk6OAscDjgNOhk4p TtVOt5xVnaOcVzpfcCG4+Lgscjnu8tHV0bXA9ZDrX26Wbllue92eTzGZwp2ya8pjd313lvsO 914PpkeKx3aPXk89T5ZnrecjLwMvjtdur2feZt6Z3vu8X/pY+wh8jvq893X1XeDb5of5BfqV +XX5q/jH+W/xfxigH5Ae0BAwFOgQOC+wLYgQFBq0NuhWsHYwO7g+eCjEKWRBSEcoNTQmdEvo ozDzMEFYazgaHhK+Lvx+hFEEP6I5EkQGR66LfBBlEpUf9dtU4tSoqVVTn0bbRM+PPh9Dj5kV szfmXaxP7OrYe3GmcaK49njF+Onx9fHvE/wSKhJ6EycnLki8nKSZxEtqSSYlxyfvTh6e5j9t w7T+6Q7TS6ffnGEyY86MizM1Z2bPPDFLcRZr1uEUQkpCyt6Uz6xIVi1rODU4tTp1iO3L3sh+ wfHirOcMcN25Fdxnae5pFWnP093T16UPZHhmVGYM8nx5W3ivMoMyt2W+z4rMqssazU7Ibswh 56TkHOOr8LP4Hbk6uXNye/Is8krzevNd8zfkDwlCBbuFiHCGsKVAFR5zOkWmop9EfYUehVWF H2bHzz48R3kOf07nXPO5K+Y+Kwoo+mUePo89r32+3vwl8/sWeC/YsRBZmLqwfZHBopJF/YsD F+9ZQlmSteT3YuviiuK3SxOWtpZolywuefxT4E8NpQqlgtJby9yWbVuOL+ct71pht2Lziq9l nLJL5dblleWfV7JXXvrZ5udNP4+uSlvVtdpxdc0a4hr+mptrPdfuqVCuKKp4vC58XdN65vqy 9W83zNpwsdK+cttGykbRxt5NYZtaNhtuXrP585aMLTeqfKoaq7WqV1S/38rZerXGq+bANu1t 5ds+bedtv70jcEdTrXFt5U7izsKdT3fF7zr/i/Mv9bs1d5fv/lLHr+vdE72no96pvn6v1t7V DWiDqGFg3/R93fv99rccsDywo5HRWH4QHBQd/OPXlF9vHgo91H7Y+fCBI0ZHqo/Sj5Y1IU1z m4aaM5p7W5Jaeo6FHGtvdWs9+pvVb3XH9Y5XnVA7sfok5WTJydFTRaeG2/LaBk+nn37cPqv9 3pnEM9c7pnZ0nQ09e+FcwLkz573Pn7rgfuH4RdeLxy45X2q+7Hi5qdOh8+jvDr8f7XLsarri dKWl26W7tWdKz8mrnldPX/O7du568PXLNyJu9NyMu3n71vRbvbc5t5/fyb7z6m7h3ZF7i+8T 7pc9UHpQ+VDrYe0/zP7R2OvYe6LPr6/zUcyje4/Zj188ET753F/ylPa08pnus/rnts+PDwQM dP8x7Y/+F3kvRgZL/1T+s/ql6csjf3n91TmUONT/SvBq9PXKNxpv6t7av20fjhp++C7n3cj7 sg8aH/Z8dP54/lPCp2cjsz+TPm/6Yval9Wvo1/ujOaOjeSwBS3IUwGBD09IAeF0HAC0Jnh26 AaAoSO9eEkGk90UJAv8JS+9nEnEEoM4LgLjFAITBM0oNbEYQU2EvPnrHegHUzm68yUSYZmcr jUWFNxjCh9HRN9oAkFoB+CIYHR3ZOjr6ZRckeweAtnzpnU8sRHi+324lRt39L8GP8k+I32zs Bfxd5wAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAAZ1pVFh0WE1MOmNvbS5hZG9iZS54bXAAAAAA ADx4OnhtcG1ldGEgeG1sbnM6eD0iYWRvYmU6bnM6bWV0YS8iIHg6eG1wdGs9IlhNUCBDb3Jl IDUuNC4wIj4KICAgPHJkZjpSREYgeG1sbnM6cmRmPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5 LzAyLzIyLXJkZi1zeW50YXgtbnMjIj4KICAgICAgPHJkZjpEZXNjcmlwdGlvbiByZGY6YWJv dXQ9IiIKICAgICAgICAgICAgeG1sbnM6ZXhpZj0iaHR0cDovL25zLmFkb2JlLmNvbS9leGlm LzEuMC8iPgogICAgICAgICA8ZXhpZjpQaXhlbFhEaW1lbnNpb24+OTcyPC9leGlmOlBpeGVs WERpbWVuc2lvbj4KICAgICAgICAgPGV4aWY6UGl4ZWxZRGltZW5zaW9uPjE2ODwvZXhpZjpQ aXhlbFlEaW1lbnNpb24+CiAgICAgIDwvcmRmOkRlc2NyaXB0aW9uPgogICA8L3JkZjpSREY+ CjwveDp4bXBtZXRhPgqcwRHqAAAAHGlET1QAAAACAAAAAAAAAFQAAAAoAAAAVAAAAFQAAEPb KmoCuAAAQABJREFUeAHsfQl8FUW2/glb2EEQHJxxQX0JA2MQBWXIk4lIBMU8QEFcAozIIItm UASREPkLhEE0CpFFEWHQABPJyOYCwgiDI4Iw+ieKAk/FyAhCZElYL1veV91d3dV97w25N7m5 uX1P/35J3+6u9Tt1qs45daoqpgQX8cUIMAKMACPACDACjAAjwAgwAowAI8AIMAI2BGJYYbbh wQ+MACPACDACjAAjwAgwAowAI8AIMAKMgIYAK8zcEBgBRoARYAQYAUaAEWAEGAFGgBFgBBgB HwiwwuwDFH7FCDACjAAjwAgwAowAI8AIMAKMACPACMTsPXSS1zBzO2AEGAFGgBFgBBgBRoAR YAQYAUaAEWAEHAiwwuwAhB8ZAUaAEWAEGAFGgBFgBBgBRoARYAQYAYFAzNHiyN4le/9PP2qU bPHrK5mijAAj4ECA+cMBCD8yAiFEgPkthOBWgaSZvlWACCEsAtM3hOBWgaSZvlWACBFcBFaY I5h4XHRG4GII8ABxMYT4OyNQcQgwv1UcllUxJaZvVaRKxZWJ6VtxWFbFlJi+VZEqkVMmVpgj h1ZcUkYgYAR4gAgYMo7ACASNAPNb0NBFRESmb0SQKehCMn2Dhi4iIjJ9I4JMVbaQrDBXWdJw wRiB8iPAA0T5MeQUGIGyIsD8VlakIjMc0zcy6VbWUjN9y4pUZIZj+kYm3apKqVlhriqU4HIw AiFAgAeIEIDKSTICfhBgfvMDjEteM31dQkg/1WD6+gHGJa+Zvi4hZJiqwQpzmIDnbBmBykCA B4jKQJnzYAR0BJjf3N0SmL5MX3cj4O7aMf+6m76hrh0rzKFGmNNnBMKIAA8QYQSfs446BJjf 3E1ypi/T190IuLt2zL/upm+oa8cKc6gR5vQZgTAiwANEGMHnrKMOAeY3d5Oc6cv0dTcC7q4d 86+76Rvq2rHCHGqEOX1GIIwI8AARRvA566hDgPnN3SRn+jJ93Y2Au2vH/Otu+oa6dqwwhxph Tp8RCCMCPECEEXzOOuoQYH5zN8mZvkxfdyPg7tox/7qbvqGuHSvMoUaY02cEwogADxBhBJ+z jjoEmN/cTXKmL9PX3Qi4u3bMv+6mb6hrxwpzqBHm9BmBMCLAA0QYweesow4B5jd3k5zpy/R1 NwLurh3zr7vpG+rascIcaoQ5fUYgjAjwABFG8DnrqEOA+c3dJGf6Mn3djYC7a8f86276hrp2 rDCHGmFOnxEIIwI8QIQRfM466hBgfnM3yZm+TF93I+Du2jH/upu+oa4dK8yhRpjTZwTCiAAP EGEEn7OOOgSY39xNcqYv09fdCLi7dsy/7qZvqGvHCnOoEeb0GYEwIsADRBjB56yjDgHmN3eT nOnL9HU3Au6uHfOvu+kb6tqxwhxqhDl9RiCMCPAAEUbwOeuoQ4D5zd0kZ/oyfd2NgLtrx/zr bvqGunasMIcaYU6fEQgjAjxAhBF8zjrqEGB+czfJmb5MX3cj4O7aMf+6m76hrh0rzKFGmNNn BMKIAA8QYQSfs446BJjf3E1ypi/T190IuLt2zL/upm+oaxf1CnPx/t301Y5dVFjsIaou4I6l Zi3j6abWcRRbI9Twc/qMQGgRCGiAOOehwoICOl09FoUCP9S7iq5oJn77vgp/2E2nwS9m2EtO 096CA+CjWMSuQ9dd3dx3xKr6tvggfVt4imKrX7zuWhXOFZn11auEeLGX0RUtGjlqCFx/AK4a VkbI83XoiquaV14fE2jdjBpIGnvOE8p7VeWV14FgpDwGxG+yUiZtjBfnPVS7WRw1aygDGHev 9ibeN6Yrrna2N0c8fqwwBIKirzN30Pvz7Vto739Ef6F/bHZdAv0OMkdDXzKHs33Y0qtDtevF UmzDRtTQR1ct+ZfAv41+dRU1rGeLrD14Dh+kvcXo97Q+H/1Xs+htT0HR9wTGjZ/FuIGREP36 ddf6Hvf0MUCEIWqOvtQnrb3J4543inwhcKrU8c9AMSj6KhQIVF8oLiyggydMNldSws9Y8G5s LDUE73rpGl59fRWWLey1cvVT9CrMhVtpyoBu9PrX/uibQOPemEt/6hrnI4CHPn/nRXp2RwL9 PSNFjnk+wpXtlWf/Vpo56EW6bs6b1PNqH6Ne2ZLhUIyAFwIBDRAn8unu1km0w0xlFH1WkE7N zGflx+GN1LJdL+tF0lzaPu40tb0jzXiXTO/uzqU2ATbnwi/y6Olxu2jcqnS6zpfwaOVY4b++ XdKPkseu1dNtnU07P0gtlbc9u3KolVlfWZxU2vhdNl2hlt2JlRH0uQ/304D4AAGS2QR4D7Ru WvKefBoUl0TrZXk/QHlbV055A6xelQkeEL8ZpbbRRtakx0LaMztFPmn34k+zqO39mbZ3RMHx mSMRfiwjAsHQ10waQvD6NzJo0JQc85X9h2+Zw2f7sEfUntr0mEbTXxhM10ml2MG/bZ7bTO/+ 0SnPeGjFnS1opJSDytDv+cjaNa+Coa+TPo/l7KRRtzqUZs9u9KUdK7QvDedYGQzBneNlZY5/ srzB0FeLG5S+4KFc8NZYyVuyED7u/cYtp+ce7WzKG06s9ChVU7bwUR3XvopOhfnwVhrUrpvZ eZVG3UFzvqCMu64ygxR+vYqevXMgrRZvoCTsXNjHbORmoLL+OHeQVs8eTcOyVmkxpkKA7ldJ AnRZi8jhIhuBQAeI1WOb0LAlVp2nrd5PPWDEqVvHeid+7V2RRp3TLMEvbcEe+mOztXTj3UOM gCmU98VC+i30q1gIcJrzhj0J+1NhPs2ckERZ74nXKVC2FwasbNsTDPzp23fSKPkJo07d5tLn L4O3q5FX3WXKnu/yqFUXWV/5lmja6sPUA11G3br6O9+KDtHEFfvp3v+KpRrAthbyCeUVaN20 spzbTWOv7Ui5RsHM8oKmtWqEsrSRm3ag/CZqaqONWfURtOrLSXQtsK6DP3Ftyk6ih7Ly9Qfz fwqtAp9dgzC1wWchbkZmrtH6Ixj6alid2E1ZrTvSzDIA1/qJ5fTeyM5mSN/tw/zs+DEKBrt0 3WDn4N9kKMzT+8ZR9Zrok2vJaFCYh0Nh1vpdvCtDvydjuvEeDH296ZMKnsym3zZRxj0HLWRf Kpy5vGYXLwZsFRgrL1ZEX9+d46XEoDLGP1meYOhLQesLDt6ShfB3H5RLeyYka1+dWMkoVVG2 kGWLhntUKsw7/tqP7p5gzCRpVE6m4aNT6PJaRfTR0gz6aLdK+lT6Ozq/G9D5VYPLks1i1GMu bcvsQ3SBqGFTIoxDAV1OK1Lmyv105+WxVA2CdiNpJQ4oRQ7MCNgRCHSAKP5iFrXtlWEmkjh6 A83ol0B1Mc1s6cxFlDuwJY3dIIOl0tJN2fTrfXnUqY9UIPV3LSGYxTYmqn8R5rBb6aEwb1tI LZB87UugeFaScmYTfDrNhYLSh4Rc2QB1N+VLWWXc7YNaAt7oykzPzM2U0S2O6qBPqAsNZtM0 KDqznIoOUUbufuoJjag6gG1cX0k4BD8DrZtWBIeQNzlvP3WH8aR6bZS3QQgK6YIkA+U3UWUb beLQjnbrbUXDu2UsNb5UCN4HaeZVrShLw8hqa8K4tHTTQiorn7kA4rBWIRj6Ejn7S70KiQ+O og4Niba+mkWfOGr17Dv7aeBNkAXw3tY+tHAJlNjpMvrllwO0y2gravQuEzfTKw/Eoe+xG7y6 pG+mab3jNC2uKeQZ/XII9WXo92RMN96Doa83fYjihy6nJcM70yXSu93Rl5p9fxB9aVUYK4Oh vX28rNzxT5Y3GPqWR1+wGaNEIeKSKfHSA+DdfPCuLJV1z1y6n/rcHEslNmO81d9XRdnCKr37 f0WhwuwcILJpfXYqWfKfh1ZntqDxyyzii86tV6tYatIYcQfCGrvB+NYpG0J1KtWKgVANoUYV qj3FRVTs8WBdC0yINWpTw7ribqUpfjkV5jGL99N9cRUkQJ8jKsbaJ7FeRpQhNrY2ZvpQBr6i CoGABwjHwE40itZsSafLwCCmEQcuxnfDHXuHRPLh5bRthD4jUg1tvD6mk4tP4qMxrexTYT5R RIUnwR8Nm2tr775dMhDu0LqnhVACFm1cSPEwHPmMK/Mt693BCwReaOiDF2yCjxQcwdsNwdu+ 9H1LALAGNK1IXaFsT+2jzTA3qneQsqDozPRRVlNoUhVmrPMqPlIE0xx4FvjF1sP6pouxrQdY ij0YcMXGYi2jj8WMzrptgzGALiBOtSK6gP6hob98gJ1Y/3j4CFGJQU9/Cr7Z5/kpg1ZA+c+g v3iMrYs2cBEDoZm26E/90E8mHc57wPyGwpq0icfDLqv0QvGZdhcML8JodARLINorSyDMYJbC XBuGqXqyoZrtCHBVR5tocrFGZCRoxvOxLlZ8w7gm2qbPdXdmmfQfHtDYAx4XLVPyuSOI70dH 3+A7UHjeBkNfT/48apUyRilwKs15/wXq0FzSxENb3xxCw7Jl/4egvRfStmdTqDEU2z2q54vo G2EgiZfCBta771r3Ij2UrptStEwMuaRRw900EW7A0kNEKswx6KOboE3pl1MeMgyFpfR7MqYb 78HQ1+RfByDC4NXrhlidJx3jqs++X8QvQ78Y0rHSUQftUeP705Alwcnof5s1kVYAX4Ht74oL IYMacazxUg/jFwN7EhX6FDh9nfwRoL6gem9ImULW6Aw8TGfeR+MXW8b0RBi1ZtyLPv9oHrXR vNcqULaQ+fI9aARYYY4bQfOnj6cEY/CqBqHw7A95NKD3Uvp118ugSDeiLiPGU+LlBTTllo70 ji+oMbh9OT2FxCTR3k/zaMrEIbTax7qFfuNysU4hGeIGhKQVGZScNstXajTl3f30wKVQSjr2 M5WS6f/EbBRmd/QLSv3YFqbrbL+XN9PUe2A5FhfWSW1aOY+mPJFpxtU/4H/rFJr+7ETq+XvL xdz8xj9ciUDgA4T3jKgY+O+8FgYjY1ZiL9puZ6XtSkPPGfCNNcM8gpZumUQtwU+1i1bR724f qOPbOpWe63cZTZhgCXi33Uq0/mNf8I+glbsm0a/+lUk3PyLDj6C1302y1jdDEJkCt+HXjejT P9JnbfVH8MI7fngBaz/HvTHJtkeBTfCRg1spgqMpALTCoLZTH/TikfEu0mfXr4ES2OTsWig6 /fDWGvj0MA4Le8xuWjErk0bOUoRmo04EzObPfIFuAw1sF9w8c1/KoLHzVG8ZESKFpuakU79b jT4Bb2x16zqN5vQmmj5ijKqfUfdRC+n5tBRqKDNxCHlyhrlG7G76y/VSEB9F727uS5smdaQp 78mI4p5M01fNpZ4JDuHKT5nbPID1l+Ow/tLMXE/rW/SnWT770wR67OWJ9Ng91rovNfdw/Q6G 30zayIYh24pieDnzqcEDrVEz29jiVJg9tGnJZHporGNsaY1xbuZTaEMKPTCL0VITyojaDJ5L 03sSTUkZYluq1B1r66Zjbd2OJZl071jJgzq6j2VvoFE9Rbu2X9+uy6GsrDSvMfC2wdk07s+p dhr/sIpa/sF/39AOSX9hJN9mcC69m5Fsy2wHDG13m4Y2R99gC1kxD8HQ12uZy7LD1OUKqzwx mEaOiTlIs9q1ojfEa3gZpPacSCP7dda8T375UFkq4lCYY9A/lZQUUc6DLWm6nLEy2k0s+pRp N0k+JWKF2cLc369g6GvyrzPRuEm0PncEXYlxs5qjL/VSFsvUL0J5g9w3cokzI/E8gl4He/5p lMH3SdNo+8LBZl++KbsflnMY40RSNr6lWt9UDyjw2B7JY4fzKXdeFo31GpOSMb5Mso0vogSr J2E51zzxi6jfqGnU7N9jaOYG7RH/RtB7H7alHncMkS8sD6tz+TS1Y5Jp2BEB7OO4GaXcPwKn r0NhLqu+cEUsvOM8tC5NWe4gZQqjFhrvnthKf+7czfQwkTx6/uc8ukUsb6sI2SLE3mvlJkoE JRCFCrNQCJrARdJOpfiuo+hPD9xKba9ri919daGi6BfonyVGuPO76WkozP+wR9OfEudS/uI+ VOJzUxZHhOHLac/Tne0CrCPIhGVQmButsq2PtK9vtjNx96mbaQ5csMS1HmtQB/nsUK1Mxi3f Q39qpwhO1if+5TIEAh8gvD0f4tPW0KIBHaguXIzrVMOyBHhZWO7Yo2glZqAvh2J85kcozPfI ATGFVn69kFpj9vTcHrnW11Ia2wDnHQbWvX5PtPxTX8Cn0NtfLaRL1mL2+QmpSDrWN0MQGQaF WdtTAEmofOJ7zac9nxFv7KSnujbXXtoEHzm4lUVhtiepPWkC0XXYd/b/Z1KiqezbA1pC00HK aSPdbe1h1Kf5mw/TbS2MN4UbaRBmHNerARy/u03dQK8+oCs0Zt1MpcwR2HhsPXINvfdEB/3J ge1fsOlXX+EBQ3bMfaekv83ZdpgSmxkhsBZsGPaOkLTyjpdMizbnUiejjoVrMujmIY6O2hGp 9TiU91GjvI5v4XgMht9M2ngVWJ9NbAVPi+0z0a8bwqg9mKowF9HfB7ekCRvsIdSnJ3N20uPG hkSmwUcNEODvse/soUdvkmOJh9ZPauGnnDLhBJrz0Rrqbhh/rDL47htS72lDOe/sMCI7eN/m po4gDyyknVPLvwmnLKmve8D0dShKZCiztYzEa8Go1gD0FZcHs3H7TsIbQHijyQv9z6GNadRN 7q0AhXnpFrjgo781vW+Qh2o0JNl3XdhNU2/uSHlGWlIY5xlmCa73PWD6Ign//EvUM/MLeq73 VdSoBtzjldl+q+/H8pYzZe0X7XKfvfToK5Yl00O904zXCbToiw3USTNyO72ckunvX+bSjQ0R 9Bwmgq5tZxqc+2Vj8qVnHHlgyGolDVn2jMyn1iMW0jtjJL9ZZTPH9tbg6a91QzJ1zqaPn6lN t94p5QMYjPMOU88rCujVW9qR2rV1n7oB8izihuAKhr5B6Qsoe/U6HvrnGEVhlryvyhSYZGj5 BwsTyaNn9+bR73tb71UoApMtQr/cSy2b239HpcJM+2HV7mhYtX1SOAEWssH0QI++9Bu4NwpL EEFhnozBZ5mv8Jhh3pR+C+W0b0Wzle+padPo8jP5NO3VHOstrI4bYHU8+XEG3aXM0lkBiMRM ThdapczWWbNRNTC4NqpndU4inmSymFNb6albrc3M7hmeTX9ocyV59n1MYzLV2YF0WvflKLpW dJh8uRqBYAYIwmZ0WdeqbsS6UnxlA7Q9j90dOzENa5wH6IPb+f2wimKGSr8sQb7a4Ty6UbEs 2wEfSqN6vUpZy+1v9Sd9lprWpFHfZyUPWelqrqpQ3tSNqaQgUgPlvCFRdWFNoZHpyXQGm/bN XmbNyApjwJLBHbS1ZjbBRwqd6uDmKKIl7IsPydSl0wH6aJMuIMi135/NakLDF+gRe6I/oNVj aIUxEyTLevSLTOr6J4U/E0fQmKRf0a71GbRik5XpSGys9vAtEKjr+VgTmZhKw+OIZi+QOOll WvhJLiX+xuHWiXCAjRIfSqeOdfIpa540Rog4qZS3LZt+ByU31iHsy/LaZ5hFHHEl0+NP3Uw7 Xsykj/QX2n/hVvwSvF8aNbD3WeJjq3vSqXujApquljkxm/69IJWaxBbR63e2pCnGbOpv+6RT aqcO1OT8j7R0dJotj9c2HKaklvYlMUoRKvVnMPxmb3cplPjLKvrEaCOa90b8zzT7pnY0X6tJ AtpxX/ogM8PwDrD4Yd+GNPqfpxT6x6fSyG6X0QfZWYonQTLlbsulm0Ffe/vVYYpPHkUpVx+g F19X0jEQ/MPgdEo4/Bm98o7FP0IB3JzVh5rBHfwnGDg6Owwcib1TYRnLMeujJ4VNkb5BG8NY 5qsMRna4jaC8xfHU58E089XwV/fQoJsbUWMY76o7hE3BH6nX68q71jfUMKNV2I+A6eswOkk3 e1Ggmqi/r6UIRwq1bVG0MsdAMf4FdLUUZhgclq2gtqg/QSA/d6SAVs/oZvfueDCXtj2JmXgo zM9DZlmqpaTICcCFXbINUBy3gOmL+Db+daQn+sX562EEvHI3jVQMu7Iv9VKqEN9/v9iXPp7g f4Z56abBtKJTO5KcO2jWThresTnVL1lLrTQvJ6twGq/A+6dmUR61NbxMhBfUnDUbqEPj3TQd k0MyHT1WMvWEV9IKZewU73s9v5my7o/T9vfxWq9rZUc9X9hJT7fcaJNpJ+duJvpbR9vyx8S0 5ZAn9OVdIrou7yoJlfNnMPQNSl8QZQd/rh+tKMxx6bRy3mN06YXTVLORh44WbKSsXvBGVeqU +tJOGtm5OdkV5uBlC3/Lp5Qs+WcACESnwgyACj+dRzffP+aiUD02dzONwgY++gWhbwAY4J/G o1zDLB7PYz3CX1+hjd98TB9ugNKba1nRD23JpG4jpEBsrEHCZg91fsmhNsqxNNK1VSRnd2+1 FGadAezCp1SYz/6QQ7/vI4WLBBozay71ah9HtcTs3+48Gj9/OyXc0pbir0igtu3iqDYG7EtY aRZwu/YKaoAAGs6NLrQBHrOLJ7faheLJcC/sbrgXVsO6mxu7DjGwtAR5OpRH7bvJ9+LzCFq0 bjy1pALavq8xdWjdnPYsG0h9M6XiBh4x1jCL0Hve960wazMscDv0pTBXh6tTzltv0eebcmj9 11C8N8E9vJZIDZvszEqiYQsMy7fBw7Xx7dA/FNfHgBXmETR5YmMa/2ymnomI/3JnmnNLK3pL f0OT31xDO3GUnRREpNB0FLyZu2oVXGlX0fnec2l+ur7ZGGEGbTaMcLqiBAVXrG/Cpj01gWdb Bc9EbDAzY7AuZJyBm22nfhbWQjDqD2+Sw+uVuqE8wyFMDbpFn1nfA0Wnb/oso5RG/wTPgKZY A+kT21qYuUqwXD0JRsA1b42gpuhn6GQ+TeqcRCuM1GTfVOvwKkq4wzJSdhm9gaZhMzlxib6p 04NWmWe8j5n0y/NpzA1JpjDRc/RyGtm7MzUQNEQe0yfOo4atb6ar4+Op7fUdqCmMOdZGRlqy YfkXDL/ZBO7e02jyZWtp/Ku6UqrRvHMB+Ee49YtrFM1/M54GDZB4GXxWvYBmYqbmr1oY/Euc RutnDNb35nDQRAhlT8Krou4v0vNDj6S2o/x5/WiQUQbxtUs66NVb0MtDb8PNcNomPY6czaxd o4Dm3GDNVAlFYUbem5SoLSPyXqMrZt4mYuYtFq6H9p3mnX0D0XzVEC3W9qZjVgsKekGe6o6N ZRBbsrWZVyrFyGWUOuhbwPR1GJ0kP4gyOvc9kYU6cwriBNy05e7otvYhA5Vy14wK7WE44Bnm UlDy/Slg+iIZG30enEuL2u2ih0ZLeQ8BHlxO/37qKpp2YzvT7Vj2/dULV1E7nLoir4v2izAM Hljhe78PkcbWeRjbXjXGNsNwUowxu8sw2b/rOUmvsZ9WD6Se441xNw59xuLBdAaG2m6jrfCJ Q3PphcHJ2h49Z/ZupNG9e5kuxGIsX7l1EsU399AH6npdZJOauYFGdo2nPV9vodhrO9OlB9Ul W3o51P+tHs6l+SP0fMR74QnR+JKK3fk/GPqKsgStLzgwEWn5v2CwWCcMFlgWapthDl62YIXZ P9rBfIlahVmAdf6Xg7T2vaW0NMe5M7YdyueW7aEBNwrLNRTVYVCY3ze+SxcLGdwYqI//THT6 5EH66YedtOuLj+mDNdgFc5cMZCkSNYtV656lFIuQwSjMMUc30k1d1Vk1Pc/45FS6u1My3dy+ M13bQrfAy9LUxUwDZGO+XIpAsAMEqWsLgU38ULhlD06g959qQc9uMMASa7QWj9CEcjETUu+I Kvxa7TwGCt5NioI3CAL7cFhR1cufUizC+PumbXLkT2FGoxY7T8OYS0eOEh07WkB7vs2n/G1r aTVmNHfKzCUPo/xHMJNjHisVhMK8KC+ZZveRwgQ29lnQmaY/PMSY3YPSvr4/rbjNstybQpNR 1lNF0AVPeejQvgL6HmXdumkpzVcs+lLYPvJ5FiUPMRRzYwYjAQqjdsV4aPe2LVTSvBVd3rS5 6e7500dp1HOMoaordBNxYn4CfXo6FDAopnUcmwaZ5XUozKqhT6S35pkmlG5MQsoy2415MCgu 3kmJaAJnzsdSrVPbaVivXiZN9JkRopybWtF0kaB6xafQ4B7J1LHjrdTmmqtsm7HVAgYNYIgM 5xUMv9kEbhhMlnY7QH2HZujV6JpNc7oV0DApgPfOpfUDTtNtvaWQbfAZdmgf2ymJ1hmV7zkR wuqtV9GZMx6qVesUrbytHb0sgZGb9NkMWcbskpi5xOXkOXWjqR8+GEh9MgwhW/JJ0UYacHsv c3l16gtf0MjbrtITw3+xRnc26PmGfGPwXW2H94nvvgHGtGeN/Iw1vL9tUESLrm9p7Bou+6cO WurV4NEcKkNwwPR1KszGRm52t2gJiu+7rX34DmK+TYTHzAwsn9GuEswwd+AZZhOcMvwImL5I 00YftOttU5Ppbawpn2Z4iYhsM15dTgeG9qK5RhlkX3r+Pzl0Sy85yVGWfrE5HYSL/l2qi76x S75I+szueTA+jjFyGUFrtk2iH2Y0oUffMl7JG5TjTYv708rUFjTVGAyl98NWGMuGmcYya8mV jHr4s0y6Y7hlEJiMTXHvbEX08ThlNlVsFrotnYzuRIvqlGllevpdL6sML+SJxk0qVlkW+QRD X1nOoPSFABTmka/upNT2ulx0FmPy780xGUbEcsoWsg58Lx8CUa0wC+hOFUNI9aCjOVlEP323 nbZuybO7UCOMcK98JTWBLhG7ZKsMIIUFaS0W5yrPe4Vm/GWWKfx5k8dSJJwzb7ITFce21IE1 TrW8m9804dpeDimUxtTw0MfPo9Na4p2r9SaFXng7m26Pa0Q10CnFQqkI9RmwVt78q7IRCH6A sLvEEqXT0vc704y7utG/jErIAVY8Ctep2rbZIqudOxVmuXmUiBcLJUes2PsaO8H2esZQ6MQ6 PXlUDuw7P7+rKLLKt4sqzKfyaeHMLPp/f5XCtsjRcSk8bBNElPcX3yVbpIkZro3j6ZPOPpQ8 8bkrZsYyE2gSZgHl7KvK02e/X0VzXsmiBR8aswMijuOSfP4DZgX6yFkBBQ8hZDSCkIEbnTkO IwFmquRVgI2D7h1n4CuNBPhYDYqxfaZRwb4BlqHEWzPJZnkdCrN8L/Kq29RDHz5mCU6+yyxL 5fsuZ9L3woDRW3Ux9hF8+IubafDtOFsWla6JvrEO6hPOKxh+swvcC7EPRSMaCsOnaWNVKiRc GzPabKT2d9kNHPYj3ZQIvn4aXhXCUGIJZdYMrbcbsNUmhKJn81aQfPKfVdS+l1TiiVQjinQ9 Xo9Ngcx12Ea8mP+gDHBLlJevvuH0wY30wK3WrvzD4TUx6NefUfs75Ky7voxJHHkmrlC5Y4u0 A6avB+v9sXZ1tYiMSy7V8Kkwix2SPbWxC7FeDz2GQyGTL9U7Ngnr0iaZuvf+E3WBt468qmHd 7PM3WPwredGet12WkB4D4uQPf/2eTN+N94DpCxBs/Cv5Aft5tDf38/BGSvaZP64dSPc8U8r4 pEQ1+0V4QvV+2vdYWd9xlNjkNzfQvgFJtqWCepIpNGfxI7T0wV7m8pZp8Ozpgubz0UtNwL9G xnE4CWYxToLBo+gX6mN8qeVY0qjXBfGebUFj3jfiKWNMTE2MS/BkLP4SHmj/Y/G6EdK8CcVb 29sA3hXBHNNqJlTKj2DoqyZXLn1BTcj4HY9jplolJtN9fftjlt7i+5LCPOpgrvcun2wR6iMr fVTLta+iT2HGZjlTJrxBe09g7c+GfJqDjXTugFnrxAm4QZzDjpOC1Ee30qSu3UzBVg4idWpj Ef/TljAo3+uDSxEtwZq7CV872krnFOq6cZVp+VfPzXQqErITFW4UdQ4EozBjbRIs758smkeT 4BrqS+DSS6ds+uAoLj+6C4HyDBA2QcALFvuMVH14KpDt7EBFyHbMMMt2LpQ1eU6lPS8rrlCK 968KQmE+vZWe/G9rPb8s/h/uS6V/vi2FDbyVAg4ExEJY7u+UlnvlvT/B0b7+EmXGZjzHXk2i QdLdW2aKu3BBzejmoTHtO5oCisThJGYFOpuzAlak5L4ptHapJUxJgdeuMFuKTiyEkvrWmEvH YQwU2nNNCCx7VaODrBs+l0a32lCYJwWgMOvnM9sFcLPMK6HkT7TqYtXS+5c0xIj2cSg/h2ZN TKPlymyNM8ZYbGL4aBXZxDAYfrO1fWOGan77lj6EXCiGYgkEtnFqb24Io/PKr/fmwBXfmqly YmR7NuhPjk36pJGqFoyoBz7wzXPCFfpnlR9lWgVwt7zXEoalwiyE5SaIIy5fCjP8qm3xJE+o fYPm2aUYquPhuvn0ddj0zlxGkI7ZrFH6bBbaeyhd8wOmLzZVysKmStaxcvrM26UwPFjriAU6 4BvjyMo2Sal03z39qVePDtQQ4WztAx4li7AmVhy5Z16os/PSZujgITIO62ZzjY+Sr0pVmKWi wwqzE1K/zzb6yL4V+O1ZnER9s3wbQGU7//E9KMwTAusXC7Cnx73pcgyzj5XiWDkbn6mlfmgu zbh6I/05U8ZVP8JbbJvwFvPQmiktKP0d45uiMNeBnFxXHAzuR2Fem96CnlljxJM44FF6MdrH SyOcejM8ny6DLAEROCRXwPxbkfqCcHl/y1gmI2vng3eFh0xdKMzWhFnwsgW7ZEugK+YedQqz 8+zj257bQPP/mGCiefYMZmdOF9GcPi1pltQ4DeaPrY4Z3PGKwqwMLrGHcuh6ZT3ygIzl9FBy Z2oqBraDsL7fJa3vVgdXmsLcGOvL1N3zpOVdZwD77J8USs2BEOc/f/dtEZ33/Ezbv/yM8jeu pZx1ykYtKFIGznzu0xbnaUI44su9CAQ8QKhQOM9bVr8p6/dlu7MPiP7bubQkqxt62IQOZdZU KMyH31ePsbIURG2G+Zh9EzLfgkgCTXhtLnW7QV/PfwjrebuNNoQGObBXkMJ86bYsum1EpoqU 9nsydgTtfsVuGoPNVD4yvuplJfrwiRY07mPjZasR9Grm49T+NzD1YzD9KBPW/mX6N8nnh5HH HUOtPGaswU7UEGY0PGoKhbUbdj29lXreHE9tfp9CneIb2YVupc7CGHDBj6EjYIUZUk7j+r4V 5uNfz6KkAYabMaozAycBiDKju9WuY8fhj14rFi7kteFCrGv9Wrtq4KED3xVR0blT9P2OrfTl F9i0bbFdwEzE+tpX7kugRmgrQp4L5xUMv9naPmgjzsi2rbM3K2S0faxva99HKqcGn3lgILqt G200wg566Qsanvgrgke2dnng738GvhwN6gBfcbA32tYZm7Jq8WttrBvcv9K3wuxlwJJtCTPF 7ZWZYumSrRtRRBEcO/UaY6dTYfbVN4jYxaWcQCEVQRFO7VPEc0VfwdB307QknMphKU7C7XzE bc3NY/q0MjqWwBBNou0FI7Sjf2ztQ+kbfdUtBv1YDZC3oVii4ZjdJqPPjq2hKOsOl3EZhmeY r/QFr893NvpIfhAGB2yelQ6Dhdx0TY0sx6mTu2fBWBpYv/jD6jR4GEmlV+Fb9H9CYfbHK73g nTLmyg99Gtak54Mw3KyFl+IzZqEtl+wGUGRhw6TCj3HEXariko1NartfjXEMCjMOLdAvRTaW Bme7fIBgndJp8m1HaXzmLCMS/LTAG091a06NQiSXBsq/IdEXzNraf4jj5WphDK0PncGOFWgM Y3zgsgVOtNDGZHs+/BQ8AlGnMJNj10oB3aDMNTSoZwf6tRhkiuFWvWQ0DZtiCWVmZ4L1gRvg dvLUewbgUmm4ALfS43l0vbnbIKzAm2AFFr0LLvuaEKuD81KYocT2jEMjR7zG9ewb7vTE2qeM u+B6CAY4++UsbFhmdbJSkN63NoP+5xnZ+WAdzZZcukZasA6upYfu6mfOOssO2zkrpZeY/7sF gUAHCHu9rVkP+3t9YBO7OYpLzEg1QLv06uQNt2p7O7fav9qZ24QOCIXapl8Q/GpDqaqHmTB5 XqzITwrVQnjfPkc5XxLfZLv+cW0aXN0MoUJbrzVYG+zF5nzz+2M3ezljqQg4FTHD3BIKfHuv fQQMRQdbU0+GwrxcVAKXqTArgoa66zjOk6GHulmuuZLPY7Ab+U3mbuRICOte14/FTsVQfs9j A622pisXUXec0T7nnriwK8wxh1ZhHbs0GhLZNrdRNiqLj0uh4c/PpUScEmDNohJNfHs/9bgm VvcAAg1nYzO1+RqK1mZooVy7amR10Vsw/GZr+4ageQzrBLsp6wS1jPFt29Q+jv0tDH5ybvqF tc/rp6bqm36dgadAp46ax5RwAbx96DQa1Bnrm/0pzOCr0rw6bN8k/1wooNk3y528RWn1Tb86 t4qlxjB6bHptCD2kjKlSofZXBrVv0OruVOy0l+Kf3dNFzmaZnyv4RzD0dc7IiSKlTlxDjz/U gUQPuvfrPJrgOP9aW8v9h+ZUHcrtPrjg9njSW0HSxm4oSBA/tKual7WoiN50eL11GbqQxj50 B/3XlTjyrljssD2QRs6zlHmz/xEKH/oTJB9VVzD0tfGv5AfgJzZ1O+3H0CPHqerF2PSrSwD9 Is73dSrMtrFSEOxEPg1rbW2YKAmoeadclk9/xl4Hn8iXxl1OyIjH09sz6L8fkXIk+ldj06+m zeCBuX8jjezYy1xioG36tWUSjpXEUhxlHJNjlboBn10+EOMfjpW61nGGuFjatCmbrv+1vlTL UcxyPwZM3wrVF7ARKIyhQi2oD7lG5S0n79qx0hXmwGULVpjL3WAcCUSfwgwAnDsAOzDxehz5 Jo6raI3FlFCYP32xBT2+xBEE5zBvm3Aaa6oUl7hOo7D5xq30w4fP0svvWAOS6pJdG1v6/+72 IY7EiCYs309/bI2NXpS1TyJQl4cn0Q01P6OX5lrKvPYeu+dOw+65x+HamWRz7Uyh8a+k0rUl B2jl22m07F8yK2s3PmsGQH7ju5sQCHiAcFS+cB2syV7nCMMgtNFwCzQEAzEIeHXyF1GYa0DJ lpbkvSvUWWRZCAwUOxbS9T/lUCuVtyAkDxqdSjV2jKG578uw+l0KIk5Xt8SH52JDDWxI9coQ Wm5scqLFUASccivMWn0P0pvY1TdbLZa0tuNoukwozMaEsaUwqzPMiDdoYi51qL6bZqRn2PZC kEJITI0iWjuypXIWtsgsmXrfR7TsbdWTBEeabMilG9B1/YhZCXONuFLnypph1sqMjXDGfm0B E987nf54A9FbEzLNzaJEPRZvzaWbqm+lR2/sRhus4DQ0fS7dAqX5q/dzKOvvVj3lrsDS00GJ Uuk/g+E3nwI3vDsG3GFtoiUq0jMTRlOc2GDfPMcyQO1Z2Y/6TrRwobhUenZgAn2zaAwtVXAf nbuHBmMTy3NQmNv42NXeaxZZmdX0+qa0pcJPsKQhTSp2OvSJ90Jp/yaHViv5W0K2/1lutW+Q RPw8O4nudbq4yvwRqDLoHwx9RfnLcia8rKfYCX0NzrbXdp3HywMfQ2GWS0UctBAziqVdhetw lrmi/JQWVnwzTz1Q+vWLxXHT92Do65N/TYMDjM7qRrEGWHKcql6niNb0DaBfFN4f8LhKSrMU Wj1Jfaxsr83MIs+xzr1ssJu1odiuGQmXa1MWFLGtJQ3Clb+JmKxRzozW00+me+8n+vvflP4F H3pic78MbXM/zEwrLtlyrCpdYcYEEc5jP7Mby0mUo+OE0SYbG4w2wbhV0Vcw9C2PvmCfYDMU ZrNt+K+db1kqUNmCFWb/CAf3JSoVZuEe9uaAVjThnxcHTZ0JEZ3JCQxenR1CAcWn0yerUmgB Opl5F08SGyrsofuw6ZbT5VFGHTEfbim3N6d/zUii/i+pyrYMYb/Lzknb9Aud1sj37N+9nuQu qfjgZcn3CswvIhmBYAYIW31hrR4Ea/V69aVxtIt4FQNNWQ5szk5e7qzrb4ZZCN+mwOc4U1XP LoHmbdpAtzc/SK9c14pe0l+W+l8KIuexAdktd3sbo7wjG7PZ9cq/hlnWd/sbSfTIHItvpaIj znJ/BgqzFDlkWX+Ei/g90kXcu4DWG3i0rM9O1dY1Njm3kXq0tytUVkD9l+kZg0fbrIRUMoyB uzSX7KewhlluWCTLW70W6oFjpbzea+5fvl2yNWUGZX4YZd7gLKjybGKFd3vXYXObsXYFTAlq /ISwtwXrV9E3V4bC5J2//U0w/OZb4D5Ii7Cr9MtK8nIWyDkrK9td7QYHafGjrWjKBiWS8yd4 dxOOZQLb0jls+tXR3InVOE4MH3SlGEfXPCENs85viru2rS156H2xx8ffnJmqzwlwx19DiZgp E9dZ26ZflvJv6xtkdHgiqJ4m4rW2o7pxPJr0dJHBQ3EPhr56OTz0Lxjr+r8oMfVXugSalreG uhgbmGnr+DHDbO7eL7xvYJgT3ms+MfJK1pfy5BVIe5E4Gjts9+ug/dYUpya+w7n5bTD09c2/ ygw91sE6+2qzLxV95qnA+sWaWK7XtvsQBxmMsbIFPA7gaXDIaeiWZ3MjlrNfNb0K8E0uafDs whraO5x52LMUewksMI+B8tg2/bJkUsv93y4fSA8rodR5aPPkFjTMnIhCH/E+xv04RT6wZx30 UzD0DV5fcCzhtPWV9hlmZ4XsWFk8H6hswfK9E9nyPUepwixAw9mQy5bSqyPTzDWFKpTxnUZQ /wGDqXv7q8zX1Wpjk6JaBfT6kHYOoUR3I2l5fCu9PqkbvfaxGQU/Uugvix6jg7O70cuGH4w8 01IMeIc3Yw1LquVeLWJKobF6rSL65+whNNLc4l98TaAnps+lDueX0oOj9HUk0l1bFxg99MXy tyjrz2O83G5EWTJmp2NtI3oi46oKR7HIsvC94hEIboCwl2P9JGxkpbjtibN9U6/Xzb+xuNWv pYe3d/I4RgkW7ZZQZGoXrYInhXQ7M3gFcewCn7fbpkhVKgmxJfk0+8kkmqPyFmbQsjMfozPL h9BTi3QFVYYXA8XZ7/PoKbg6ynWdWimTJ9G8gVfS4tSBJt+L9fw942PJNkMGN+dN4jzkUqzB /up7fGcWJaVm6qCIOmC9bnehIOBc1Odvtta0yfeirD99mEl3P2GtCxORuwxfSA+3+pH6p8n+ AQMnzqdu1dAQQk7spr9mZtBzi6QKbmSJo5cmPzHR1nf9tBHHSkm3TmfdbOsnFfo4Nv0ysYXC rJ7DbL7XFGai1dgNeZhhOZTrS01lFmV+LX0ITV1mGRS0Uvsos3A5PfXVKsp+cSDlbDLqptwe Gb+cUlNwNjPamLgi1iV7BRQiaYSVxii0u68X96MBWZK2Fj/ZlUzrvc5PRfSP2S/S4OedM1AJ NHziC5R6Vwd9aQLwOrcPCrO5ay3obpxjLNI5+RE8S4bI9mjPQ8xwJcsZLuk9YfKJh/694i2a PXMMfSSXPWjUgWvnw9k08uFUaqlsWFUHLqltTJdUpe2hDKYxzYhP5HQx1vlB2wAL+fs719iM XgE/ytuf7vl0Lc3P7uezPaeOXkj9e6dQU6M/FcWt2xR9g9hkTbYPnHtr9qs+MfJVSQ99/rd5 9NLTGT5kAoRHPzr5yVG2/kIqTr5Sc/O7YOhr845y9q0GWDalGu+cfSYF1C96KP+v9iUOIhuZ ptiYr/6prTToemvTS/W4tmqOo9zkEieRhrqkwbMvnxbMzaLnFziNPMk4Jmsa9VRkY7G532ev lNLvi8Rt4wzRNGxg2OUKY9IGRgPVqCCMvdkPYpa5go02wdBXFD04fcFDGzOVmX5prDT7Sj1l X/8rUrbgXbJ9IRzcuyhWmHXAjv4ijpY6qB3BUqsO3EOwC02DBo2w8YwdUGFxVc+FO7IXR0Cc wBmX1RtRg8a6tVzEqIvOquiHg1R8Hms7a+FbA+ubPUV0FFDAG4t109ikq2BPEVR4nEmKvBsY eUsB8NjPRbT/KHZvqY5NW0TZDAHRmZ4plOJDEep1sriIjsldX1DOpko5tbiwRF6CAdlr6ZMz YX6OWASCHyAcVQZfHMa+TCXqa3T8jbFOy9kcz57AVgAnrYDaAI4B1fnerjDr4c8XFtF3h7zb ujZ7BFY68AM2gNLOlr0IbxnKm8jzp70Htc2latVBnLr++VEq/yePoE/AjvnaVQZB3FkvWd9j h+ByKhcYIjFzx2CkfRh5SCylFfgk6v0f1F+s3vLVB+kFEglBIALuoibncXTU0YNFdAjnN4vL 2R+Jd+psUWl1c9bDpI+f8pK/9yJTXKIPOmdUUu2bRJkL9xdRsSgzGo+vftLECuloZUZfJuuo 4dPYux8UykWdMHdm5eE3f7Q5UmitUxW4SmXGSS/Z7sSi1iMYMwqP6PiKcaMpxg0boypCm990 JPaSF/Bs5uH85uATbfzBMUnHZLtsgDPBHWOqPG4KSZWpb6BiuxJAyqyZfVdtkWJorvLQV5bo 2GGi40cNbETn6WdcV/ERcW3tA88qLcT3i12SJh70n2cgnwg+UuUNGd8p68j30XAvD31t9HHw g8DuAvq9I8ftKMq+X7wNtF8sdazE+Ccu7PVHJzF2mxf6x8boJ0Wzs5UXz+o4IcNrxxMe99Cx Y6fRZvRxuSn6XtuFNNVjoPz1+zKOs79RMSjGmHFWDoyIUBOeXw0VA5tMI9h7eegr8gxWX1Ax EeN3WY17TqwkzwcqWwSLF8ezIxD1CrOAQ2zQKhRlhU8tlNC4a2CgF+fI2a6z6PwQ74IjkhAy a0PAKIag7vwmdq+8gHzke1WIxDHQdErt2JCZKTRCACqCgH1OEb5FWYSQUBP5e1AWcanpiefj R1EvfHMUUXzSNhZzylDaB/7nKgTKO0CoYBRDATyrtEFp0FHDiN/OTl6ei3oePHFUUaRNhcyR gBAodYHO+iCVhAunwQsQOiQPyRDiDF5YnMzyyfDi+6lj+EM8Gx+Ar4VR7CziyPfSgHUSfHPK 4Cl1DZbMy3n3V195ZqMMb5bJoWia7xHwOPjcoygoIq4QZISRzKP0D1K5F99Fn1KMOp5XaCPe i0v0OQ3Qd0k9srS6+aWPv/L6e69nTWp7cfZNwoJxtBhlluAbcWSZnf2tRkNBKx/hq9WAoRKG R7H7b7iv8vCbkzbSGOXkB8lP/tqdxMAZT74XeNW7BG3KeOFMR+ySXc/A8hR44aTkBYSXeYuo zvI6N4nyN66K3WBrg151ZQGQlv+256EdX+7CgHWUVjzdi15X1kKPwd4i94m9RXCp/KC9CNG/ 8tBXLVJp7Vnwex3gUwdGRvWy4Y0PKi3UcKX9Lk0mEPFEf+HkvdLSc9u38tDXRh+ML05+EFid QP9+Wunf1b5fwzLAftEXj9vSRJ95GP2s7DZVw5KT733tGyDKdBaK/gmM277GF9h6qLFDNi61 30d6Tl5XDUNOo4IvJV7DKch/5aGvzNJfv6Z996MvOBVmX21Dpq/enTSSPB+MbKGmy7+DQ4AV ZgU3MRl7HpbXahiwLkD4FDtU1nIMWkpw7acWB2GrCUbBjDGimtcZCOlCIBQ74NVExyKFVjOA 8wc6UqH8CrlXCDWxjrzP49sZ0dkaTFnzognqGcgyak+IG4tyljGqngD/j1gEKmKACEflz0Nw 0JRm0dbBCzUNAV6WRSi6woBUZt5CRA8Gfo23wKSxirAu06wyd/C4VNjL0gfJcpuY4YXoj2pG AJ9rdBSzXShvWWip1lHUWwj4Ze0HRfhQX1WN3y5gzNAMqsBXYCyMRNUrufMXRzWe02Y0QWMf 41ppNHEe62KFlefG4g3qU1meUhVNX7X9BzquW1gE/kvQRMg40vAYKF0CzzEyYlQ0fYOttdou LtYvmn0i+NvXWBlsGbziCfkUf1qbEX2JQ971Cl8FX1QkfYPRF6ogJFykABBghTkAsDgoIxBp CFTkABFpdefyMgKVjQDzWwUjDjfsu7EWc4ctWfvGYdIzxBYkRA9M3xABW0WSZfpWEUKEqBhM 3xABGyXJssIcJYTmakYnAjxARCfdudbhQYD5raJxh0v2ug9pF9bpHz1emy69Io4S2iWYG72J 3Oo3g8dIRWfrJz2mrx9gXPKa6esSQvqpBtPXDzD8ukwIsMJcJpg4ECMQmQjwABGZdONSRyYC zG8hohuWYBzCWkznpa5/dH4LxTPTNxSoVp00mb5VhxahKAnTNxSoRk+arDBHD625plGIAA8Q UUh0rnLYEGB+Cx30YrMdsfGQ2MBI7DMSi91zK3ujN6Zv6OhbFVJm+lYFKoSuDEzf0GEbDSmz whwNVOY6Ri0CPEBELem54mFAgPktDKBXYpZM30oEOwxZMX3DAHolZsn0rUSwXZgVK8wuJCpX iRGQCPAAIZHgOyMQegSY30KPcThzYPqGE/3Q5830DT3G4cyB6RtO9CM/b1aYI5+GXANGwC8C PED4hYY/MAIVjgDzW4VDWqUSZPpWKXJUeGGYvhUOaZVKkOlbpcgRcYVhhTniSMYFZgTKjgAP EGXHikMyAuVFgPmtvAhW7fhM36pNn/KWjulbXgSrdnymb9WmT1UvHSvMVZ1CXD5GoBwI8ABR DvA4KiMQIALMbwECFmHBmb4RRrAAi8v0DRCwCAvO9I0wglWx4rLCXMUIwsVhBCoSAR4gKhJN TosRKB0B5rfS8Yn0r0zfSKdg6eVn+paOT6R/ZfpGOgXDW35WmMOLP+fOCIQUAR4gQgovJ84I 2BBgfrPB4boHpq/rSGqrENPXBofrHpi+riNppVaIFeZKhZszYwQqFwEeICoXb84tuhFgfnM3 /Zm+TF93I+Du2jH/upu+oa4dK8yhRpjTZwTCiAAPEGEEn7OOOgSY39xNcqYv09fdCLi7dsy/ 7qZvqGvHCnOoEeb0GYEwIsADRBjB56yjDgHmN3eTnOnL9HU3Au6uHfOvu+kb6tqxwhxqhDl9 RiCMCPAAEUbwOeuoQ4D5zd0kZ/oyfd2NgLtrx/zrbvqGunasMIcaYU6fEQgjAjxAhBF8zjrq EGB+czfJmb5MX3cj4O7aMf+6m76hrh0rzKFGmNNnBMKIAA8QYQSfs446BJjf3E1ypi/T190I uLt2zL/upm+oa8cKc6gR5vQZgTAiwANEGMHnrKMOAeY3d5Oc6cv0dTcC7q4d86+76Rvq2rHC HGqEOX1GIIwI8AARRvA566hDgPnN3SRn+jJ93Y2Au2vH/Otu+oa6djEluEKdCafPCDACjAAj wAgwAowAI8AIMAKMACPACEQaAqwwRxrFuLyMACPACDACjAAjwAgwAowAI8AIMAKVggArzJUC M2fCCDACjAAjwAgwAowAI8AIMAKMACMQaQiwwhxpFOPyMgKMACPACDACjAAjwAgwAowAI8AI VAoCrDBXCsycCSPACDACjAAjwAgwAowAI8AIMAKMQKQhwApzpFGMy8sIMAKMACPACDACjAAj wAgwAowAI1ApCLDCXCkwcyaMACPACDACjAAjwAgwAowAI8AIMAKRhgArzJFGMS4vI8AIMAKM ACPACDACjAAjwAgwAoxApSDACnOlwMyZMAKMACPACDACjAAjwAgwAowAI8AIRBoCrDBHGsW4 vIwAI8AIMAKMACPACDACjAAjwAgwApWCACvMlQIzZ8IIMAKMACPACDACjAAjwAgwAowAIxBp CLDCHGkU4/IyAowAI8AIMAKMACPACDACjAAjwAhUCgKsMFcKzJwJI8AIMAKMACMQWgSWLl0a 2gyqWOp9+/atYiUKbXGYvqHFN9ypM33DTQHOnxHwjwArzP6x4S+MACPACDACjEDEIMACd8SQ KqiCMn2Dgi1iIjF9I4ZUXNAoRIAV5igkOleZEWAEqjYC+/fvpwsXLmiFLCkpod/85jdVu8Bc uiqBgBS41x1Zp5XntSGvVYlyVXQhZD2jdYbZ7fVm+rrbcyLS6HvkyBES43BMTAxdcsklF+3O jh07RmfPnqXq1atTo0aNLhrejQHOnz9PRUVFWtUEZgK7SL9YYY50CnL5GQFGwDUI/Pzzz5SW lkZSoJAV69ixIy1ZsoSuvvpq+YrvjIAXArLdsMLsBY0rXlX9SNYAAAObSURBVEj6ssLsCnJ6 VYLp6wVJ2F+cOHGCEhIS6Pvvv9fKsn37du25tIINHjyY3njjDbr11ltp/fr1muJcWng3fsvL yyPZT23dupXat28f+dWE1SQqr8OHD5ccOnSoBJagMtUflhIt/NGjR8sUngMxAowAIxAIAhiQ S5o3b16CUcXnX4MGDUp27NgRSJIcNsoQePvtt0vE35DXhmh/Za3+vn37SnJyckr69+9va4P3 339/ycaNG/0ms3nz5pLExESzvT799NMl/sbILVu2lDzzzDMl7777rt/0yvpB1rOs4d0Srjz1 hsdKCYxuNvreeOONJWvXrvULD9PXLzQh+cD0DQms5Ur01KlTJVCYzT4uLi6u5OTJk6Wm+fjj j2vh77vvvhLMtJYa1q0fYfwxMYPC7IpqCjeDqLuOHz9ecs0112jETElJuWiDFgzTunVrLTws RiXnzp2LOsy4wowAIxA6BEQf06FDB3OAGT16dElhYWGJUGTGjRtnvhf9VnFxcegKwilHNAJS 4A5UYRaCnT9DjXg/atQor3FSKFoyjjDmyN+ijToFyjNnzpjt+x//+Ee5MZb1LHdCEZZAsPWG e2jJ0KFDTRpJWsn7Sy+95IUE09cLkpC/YPqGHOKAM3AqzIJnsrKySk1HKsxl0S9KTSiCP7LC HMHEU4uuMkBZLECBhlfz4t+MACPACFwMgXXr1pnC7KBBg7yUk4yMDPP7ypUrL5Ycf49SBKTA HajC/Mgjj2jta8KECdqM8v/+7/9qM4/SsCyExA8++MBEVcyaJCUlaXGEsiVmLw8ePGgalp1t dMWKFVpYIUBWhMFZ1tMsUJT8CLbeCxYsMPuPKVOmlBQUFJTs3LmzRNJd0Ff1JGD6hqdBMX3D g3tpuaryvzQwiTtcs/1GY4W5pIQVZr/NI7I+qAxQFgtQoOEjCw0uLSPACIQbATGjLAZhMVMn XLOd1969e02BVwi5QkHhixFwIiAF7kAVZuEmLZYoOS+xdEl6V82cOdP8LLy0xHuxhEAoyvKS ipkaVh0/K8o1T9ZT5hst92Dqrc7ujxkzxgaVUIx79Oih9S2PPfaY2a8wfW0wVdoD07fSoC5z Rmr/pSrMpblml0VhFobDH3/8sUQYJ7/99ltt3BeeIP4u8U1+Fx48X331Vck333xjW1YqvqsG SWEUy8/P18L98ssvXknv3r1b+y7CYP8Ur+/OF8K7TZRVprlnzx6/S3DcqDD/HwAAAP//guTa mAAAMLtJREFU7Z0JvFVT+8dXhvJveMtMMr5es4yJEhlCevGiKN5IVEpFUUKh0iiKBlNJSFQa 36REJEqTSgPJHImXJpSK89+/1ftsz1l373POPffce/e557c/n3v3sNZee63vs87a61nrWc82 sRzctmzZEqtatWrMGBO77LLLYn/88UdCCvmNnzAxBpIACZCAIqDbl5o1a8a2b9+uQnce/vnn n7HGjRvbNmu//faL/fzzz3ni8AIJjB49Ooa/5k81t3+ZIHL33Xfbejdo0CA/Oamzbl184okn bNxnnnnGj4v84F3bsGHDpO9a/6YkB1LOJNFKXHA65f7iiy8sf8hg+fLleZhMmTLFhlepUiW2 adMmG0755sFUJBco3yLBnK+HyG8Bv5/OnTvH6tSp4/+eHn300cC02rRpY+OE6ReTJ0+Ooe1E mvoP1yZMmJAnzdWrV9t4FSpUiI0bN87XX+TeOXPmxMaPH2/jHHHEEbGPPvooVrdu3bi0Effm m2+2/YsffvghrhySTseOHWNbt27N8/zPP/88JmWSuHrftGnT2Lp16+LuGzNmjP/8+fPnx4Vl 64nJ1owXJN/6BxBWoXX6yeLrDu5vv/0WW7x4cWzp0qW20q5fv14nxWMSIAESiCOwcePG2HHH HWdfLj169IgL0yfywkLHFvdwIwGXgHS4M6Uwb9u2LVatWjVbN4MUZrcuSidJ4v7yyy9+3V62 bJmb3bTPpZxpJ5ClN6ZT7kWLFln5uYMbguDTTz/1w9GRxiZ9HspXKBXNnvItGs75eYr8FqAg Pv/887EVK1bY34sojEuWLMmTnLyrXf0CA9/t27ePu1/S0XsooFqvWLVqVdw90l+QexYuXBiT tleuhe3RDgQp6xK/Q4cOceWBLiNhifa1atWKy7PODxXmOKTZdaJ/AG6FDipJovgy+o4K/t57 7wVWLIxC7dixIyhpXiMBEshxAitXrvTbDT0z52LRL6BPPvnEDeY5CdjZZXS6M6Uwi7KFjtLU qVN9wloR1rOW8j4UhXnYsGG2bmNmA51FbLCOWLNmjf2Ta37CKR6ko1ikmHSko6VT7pEjR1oZ uMqvFFR3xqVjS/kKnaLdU75FyzuVp+n+v7RrvXr18t/ZRx11VAwDi3oLU5iHDx/u34c2FfLG JBt+b6+++mpcmDwL6erfKO7DTPPQoUNj999/fwzPh3Kt+weIg4FOTN5h0u65556LSxvhF198 sQ3/6aef4sJ1O4H2+YorrvDv7dmzZ+zrr7+2s9AbNmyITZo0yQ9DmtJ+IM86P/o6wrJ14wxz AU2yW7VqFVdhUGmC/q655pq40ZdsrTDMNwmQQGYJ6Jehfkm6TymJLyC3jDwvGAHpcGdCYf7x xx9jMO/D+wx7dOz0Jp1CdMzmzZsX051BKNHoUMn9n332mX3/derUKe79iM4ZBozyu0k583tf tsdPp9zSbqAPErT8TCsE6GDLRvkKiaLbU75FxzrVJ+nfh7yfoeDKsk60j65ptvx29IQcBgql PcQ9CxYsyJMFPZuLWeC1a9faOLqPgHsxoyybTMbJ7xzhUKLd9loGMxGOpV+uko+ZZYRBGYcJ Njb93LD2A3UW9+FPK8Y6P/q65Dsb91SYC6gwyw9DKkyDBg18W36YaugfSKLZo2ysPMwzCZBA wQnol5K8kINSLYkvoKBy8lr6BKTDXVCF+ddff/VNsfFu07PIkjuY77qmgYgrnccBAwbYTpSY +D3wwAN+h6x3796x2rVr23MozbJ2VtJOtpdyJotX0sLTKbe0G2EdXq0QaDlTvkVfeyjfomee 7In696HfzxgklH4/9to0W/QCrTBrSzKEh23STiJNUTR1H0Fb6+g05HeO+15++WUdZI9ljTPC 33zzzYTh8lws/Zo+fXps8ODBsQ8//DDPPbigyyX34brOj76OsGzdqDBnUGGuV69enllk7XAD yvPmzZuzta4w3yRAAoVAQL8M9QvZfVRJfAG5ZeR5wQhIh7sgCjOUZcxAoGOFv2nTpoVmCqaA 6HyNGjXKdtJkqQBmpzFDgtkKmPCh0ynK9dy5c216uCbro7W5d+jDVICUU13KicN0yo17IEd3 jaEA++abb3xZw3mQ3ihfTaPwjynfwmec3yeEKcxIRyu32jQ7SGHW729pA4PyAsVU2l60q9h0 HyFs4k2nry1F5Bk6XM9QB4WHKbgw3/74449jb7zxhlWimzdv7ucVedb36efp6/K8bNxTYc6g whxWKbp06eJXqrA42Vh5mGcSIIGCE9AjtGEvQzxFjxDrmaCC54AplBQC0uFOV2GGoqtNDV9/ /fW00MgaP6yxwyZ1XK+Pw3XMRqOjJbPSuJbKJuVMJW5JipNOud9++23LGAMYQd71xemXNsVM xozyTUYovXDKNz1uhXlXIoU5zDQ7mcKcSA/QyrEMoOtrYX2EZApquuFYxzx27NiEjsJEwdfl Sva8wpRZYaVNhTlDCjNmj901AyK0iRMn+gpz0MiPxOOeBEgg9wjotU3oiIZt0kl1lY6w+Lye ewSkw52OwgxrKL2E6N13300LIBx6oQMVtAZPmygi8ZdeesnGhcOb/GxSzvzcUxLiplNu6WzL bL/LQZyVhinUbnzK1yWSuXPKN3MsM5VSIoUZz3BNszEDK74adHuXqgIpA1hoQ4MUZq2U6jIm Sz+dcFiYaKdfohhjD6dhffv2tY7H5LrOW7Ln6bxnyzEV5gwpzGHrg1AR5IWFSjVixIhsqRvM JwmQQBEQ0C/kMLNJOOuRNZ+pdmyLIOt8RMQISIc7vwozTASl04P6hRnhdDdxLqNnjfUMs16v 3L9/f/tcHTeV50o5U4lbkuKkU27t8Tqo/yHyQudeHAglYibxtcwo30TEUg+jfFNnVVQx9ftZ FFj32do0+9RTT/WXtIQpzO7SB52eDGChPZaBRK1DaKVU35dMQU0n/LXXXvPfCxhwe+WVV6wj MjCRDZ8LlHeHNvVO9jy5P5v2Oa8wN2zYMNBzpBai/sG4irGYXrjX9f36u235Xaul0+ExCZBA ySQg7QhePPAo7G7fffed/1JKpc1y7+d5bhCQDnd+FGbtZwP1D50kzCJipkP/YeZEK7tBRFev Xm3rqTuoA+cxMnstznHwXuUa5iCK4ddEvuExgkNEyUWnV6+fHDdunN+u4BuzyTbKNxmhgoVT vgXjVxh36/5/mMLsmmaLAqkVZhlUQlizZs1Cs6r7AmKRWlwKs7QbyHOQozAUYsiQIX4bgk9d yUaFWUhk+V6PuKbiiEt3KNxRWKncQW7aBZOuOHoERsK5JwESyG0CeiS3Y8eOcTCwhkg++YAX F759yI0EgghIhzs/CrPujElHL2yfaGYE9VQ+sxi0zk46X1Da4HW1Tp06tqOVzhIDKWcQg5J8 Ld1y4/M0MmAB2UIGWsbov+hZoyCGlG8Qlcxeo3wzyzMTqaWiMOM5rmk2fl9aYdZLrxA2a9as PNnTaaBdFJ8Duo0uyhlmac+RX/hCcDc4DMTgqLQlekBB6z1heXbTi/p5Ts4wo+G/+uqrfSEn 64DC9EgqhLvGUBRmhIuHUC10/ZJBHBkx0nF4TAIkkNsE9CAe2gk4S8KsMrxSduvWzW9/0OlN NsuX2yRzu/TS4S4shVlmh4MoQylD5ymsjsL7NmZW5F2KPeLDAiu/m5Qzv/dle/yClBvykUEK LYPWrVsnVZbBjfIt/NpD+RY+4/w+IVWFGelq02z8xrTCjPAXXnghrv0bNmxYDI4W169fH/cZ JtyrBx2LS2EeOXKkn1+sWYb5NdY1b9iwwToC0+2Im2cqzJB4CdmGDx/uV4REL+3Zs2f78VAh 3BF2rTA3bdo0zxogPWIUtj6xhCBlMUiABApAQJtsuS8inGNWiN6xCwA4B26VDnd+FOaixoKB IJh6f/nll0mXQ4XlTcoZFl5Sr2ei3PjMF/jDxB5KcKY3yjd9opRv+uwK604ozPJJPD2DGvQ8 d+DbVZgxgda5c+c4nSLoXQ+9QvsTKC6FGd9i1zPIQXnV1/CNaPhbwUaFOaiGZOk11zwCQsdI K9YYQ0nGSJA42ZEKgXXKuhKj6FphRrwmTZrYjgBmhrBgX+7FfsaMGVlKi9kmARIoCgJfeJ6K 9Wd9pP0488wzYwjjRgKJCEiHO8oKc6L8pxom5Uw1fkmJlyvlzpVyuvUyV8qdTeXUM8x6ja4r OznXk2Rh/kZgji1KuLzjsYdPh5kzZ0pS/h7vfokXtqxzypQpfpyggXUdjplidwsLh9l1gwYN /LQlH1CkMQP9+++/xxo3bmzDYV0kXwvS6SWyTHLzEeXzUsicByAnN29Gx3iVNqWyexXBLFiw wOy5555x8du2bWsGDhwYdy3oxDPrNu3atQsK4jUSIAESiCPgvSBNqVKl7B8CDj300LhwnpBA EAFvVN9enrF+ht0/1fypoGhZf03K6XXksr4s+SlArpQ7V8rpyj5Xyp0r5XTl6557Fh7Gm3W2 f6VLlzb777+/GyUy557puPGW1RjkE2pj5cqV/f5JZDJZyBnJaYUZbFEJnn32WeN9Ny0QtTeK Yvr06WMaNWpkypQpkyeOKMyI55lr2HQ+//xzPx6uv/jii8ZbO+Rf4wEJkAAJkAAJZJqAdESp MGeabDTSE/mW9IGCXCmnW6typdy5Uk5XvjzPbgI5rzCL+Dy7e+OZURvP/MLssccexluLYCpV qmT23ntviRK41wozZoUw+uKtETLlypWzaR1yyCFml112CbyXF0mABEiABEggUwSkI5qp9KKe TklXHF3+lK9LpGSdU74lS54sTckiQIW5gPLUCrPnRCOPyXYBk+ftJEACJEACJJASAXa4U8KU tZEo36wVXUoZp3xTwsRIJFAsBKgwFxA7FeYCAuTtJEACJEACJEACJEACJEACJBBRAlSYCygY KswFBMjbSYAESIAESIAESIAESIAESCCiBKgwF1AwLVq0ME8//bSBcy+aZBcQJm8nARIgARIg ARIgARIgARIggQgRoMJcQGFs27bNOvdCMhUrVixgarydBEiABEiABEiABEiABEiABEggKgSo MEdFEswHCZAACZAACZAACZAACZAACZBApAhQYY6UOJgZEiABEiABEiABEiABEiABEiCBqBCg whwVSTAfJEACJEACJEACJEACJEACJEACkSJAhTlS4mBmSIAESIAESIAESIAESIAESIAEokKA CnNUJMF8kAAJkAAJkAAJkAAJkAAJkAAJRIoAFeZIiYOZIQESIAESIAESIAESIAESIAESiAoB KsxRkQTzQQIkQAIkQAIkQAIkQAIkQAIkECkCVJgjJQ5mhgRIgARIgARIgARIgARIgARIICoE qDBHRRLMBwmQAAmQAAmQAAmQAAmQAAmQQKQIUGGOlDiYGRIgARIgARIgARIgARIgARIggagQ KPXTzxtiUckM80ECJEACJEACJEACJEACJEACJEACUSFAhTkqkmA+SIAESIAESIAESIAESIAE SIAEIkWg1OYt2znDHCmRMDMkQAIkQAIkQAIkQAIkQAIkQAJRIECFOQpSYB5IgARIgARIgARI gARIgARIgAQiR4AKc+REwgyRAAmQAAmQAAmQAAmQAAmQAAlEgQAV5ihIgXkgARIgARIgARIg ARIgARIgARKIHAEqzJETCTNEAiRAAiRAAiRAAiRAAiRAAiQQBQJUmKMgBeaBBEiABEiABEiA BEiABEiABEggcgSoMEdOJMwQCZAACZAACZAACZAACZAACZBAFAhQYY6CFJgHEiABEiABEiAB EiABEiABEiCByBGgwhw5kTBDJEACJEACJEACJEACJEACJEACUSBAhTkKUmAeSIAESIAESIAE SIAESIAESIAEIkeACnPkRMIMkQAJkAAJkAAJkAAJkAAJkAAJRIEAFeYoSIF5IAESIAESIAES IAESIAESIAESiBwBKsyREwkzRAIkQAIkQAL5JzB18vj835TFd9S97Moszn3+s0755p9ZNt1B +WaTtJjXXCNAhTnXJM7ykgAJkAAJlEgC7HCXSLH6haJ8fRQl8oDyLZFiZaFKCIGcVZg3bFhv YrGYKVWqlKlUac+k4vxl82azfcd2s+suu5q/VayYNH5xRpCyJcoDylG6TBmzxx57JIrGsAwS 2Lxpk9nxxw5TevfSplz58hlMmUmVVALbtm0z33+/1pT36stee+1dUovJcmWIgHS4S/rMa66U 060WuVLuXCkn5Zt9FiLox636dJX5cd06Kz7oEPsfeKA5/PDDTcWKlVyR+ufS//MvOAfok++2 226mbLlyTshfp3/88YfZtGmjfyE/+ojWC8qULpPwOf4DeBBHICcV5t9+/dVUr3aK+fKLLyyM OfMWmRNOPDEOjHtyW8vm5vnnhpsaNc82r02bYXbddVc3SiTO3bIly1T5ChXM/Q90M01vaWbK eAo0t8IhsHXrVlOrxhnm45UrTd16/zQvj37V7LLLLoXzMKZaYgg88nAf8+D9nW27M3X6m6wz JUayhVOQXFE0cqWcbi3JlXLnSjkp3+xRmH/77Tfz9JNDTJf77nHF5p+3u7ODaXtHO7PPPvv6 13Cg+39xASEnLW9rbVrd1tYc5inhepsw/lXT+LqG/qUj//EPM3/RUqto+xcDDlZ98ok57eQT /JAevfuatre38895kBqBnFSYUXnPO6eGWfbRR5YSKt37Hyw0//d//xdKrcOdd5gnhww2V13d wAx//sXIdlzdsoUWyAnAD3OuN3DAmU8HTIZOMZBxbq2zfIV51CtjIzvokqEiM5kCEpj51pvm 8nqX2FSi3u4UsKi8PUMECqJowOLq1bGjTcc725sff/zB5ujkk08x3Xr0Muedf0FgDhfMn2fu ubuDmTvnfRuODuNdHe4OtMJauGC+mTxpojnrrBrm4rqXBqaX6sWClDPVZ0QxXkHKTflGUaLx eaJ843lE5ey///3R1L3oAtt/SyVP7iRcuv3yKa+/Yc45t7b/SFdhRsDij1aavx95pB8n6GDI oIHm7g7t/aBH+j9umt/a0j/nQWoEqDD/j1PPPg+bNm3vCKUmCnPUZwf1D/Ogg6qYES++lEcx 22333c26tWvNCG/GfPKkCX6ZH+ja3dzVsZN/zoPMEfj999/NP+teZDuWNzS5yQwc/GRkB10y V2qmlC4B/C6vu7aBf3vU2x0/ozwoVgLpdrh37Nhh7mp/uxn2zNOB+e/Vt59p3eb2uDA9oANL JSxbwobB13kLl8QNQG/fvt3UOe8cs3DhAvOfqdPNubXPi0srvyfpljO/z4la/HTLTflGTZLB +aF8g7kU51UMNLVs0cyMfGGEn42+/fqbWueca/baey/z6y+/mlnvvG3uaHubH+5OQOl+OSJh hrdy5cp+fBz8svkX8+mqT8zjj/X3r++7735m0dJl/rLRIIV50JCnzI03NfXvcQ902ythVJiF RP72VJgVL3dUSAWZbFSY6ze41jw74gW7TluXRR+/PGqkada0ib3k/jh1PB6TAAkUPgGsM+r5 UDfzxOBBcQ+jwhyHgychBNLtcI984Xlza/ObbaoPdnvIXNOwkdm6ZasZ0L+fXYqEgGkzZtql ATj+888/7QDgu7PeMVCmb2vd1vz003/9WZjRY8fbpSeIi+21/0w21za4yl7LhHVNuuXcmZvs /Z9uuSnf7JA55Rs9Oa1b97058rCDbcagCM+YOcvsv/8BeTKKeJgUwbI7bOMm/sfUuehie6wV ZkxkLVq63JQtW9aGuf+++vJLzwK2pm/lA4tW9OWxuQozLID28CxjsVwL65+DthXLl5vqp58c F0SFOQ5HyidUmBWqRKbZqSjMWJC/du13Ztvv26ySijWqB1WpElqRMeqLTSo6RujXejO/6IzA PLzyQQf5YSqbCQ/1DzOVTrbu+CRSmFG2b77+2vzqmRbv7s1Ql69Q3hx4YOWEyjgyCgVg7Xdr 7WwqnCPss+8+oc6LXB5r1nxjNm7Y6eCggjeDcfAhhyR9HtaZfPftt6Z06dIGs7rlypfzRvIO CmWGZyJfsiYdpjc///SzjZ9KGfG8b9esMRjFgxwT3QOGGK0UeYdmKp8Bwhhpw4nbPvvsE2gS mc9kGb2ICcx+d5ZVONzH4nd5+hlncN27C4bneQik0+HWMxB3tL/LdPdMsGXD++Ga+leaaVNf My1atjIPPzLAtpeyxOSn//5k5i1a7K/ZE8VMd8j0O+md9+aaU089TZJPe59OOdN+WIRuTKfc lG+EBJgkK5RvEkDFEKzX/yabzdVWN9oqR7eBifrZUrzRr4wyNze5wZ7qttRVmCV+IrPsAY/2 y7PuWqcpaXCfnEDOK8wdO91rsA7rrTdnWFq6kmt8yRTm11+bYlrd2twfFZJ78eMYOPgJU++y y+WS3X/91Vfm+GN2rjt4653ZZtHChdYkLi6SdzL4iadN4xubJFUU5T79w0xFYYaSBScCEyeM M2E/5Pffm21aNGvqO0mTZyF+74f7mQbXNMyTv2+/XWO6d30wzoxF7ruwzkWmR6++5rjjj5dL RvPATMYb06eZfn17++E4OO20081TQ4ebo485Ju46TqAcDx821FoCuIG47zFPBiedFD/KJrMe GDV8dfxk89iAR/zZFEkDZXx+5Chzdq1z5JLdJ3IAgXueeOqZuHV6Wi4YLRz23PNm0sTxvgMH d0ZGHgb5tG7VwubLlQ/y0KdXD/Nov74S3d+jznS6t7M55JBD/Ws8iDaB8ePGmhuub2QzWe2M 6gYv51YtbrFmrKn8lqNdOuauKAik0+HWbS8cyBxz7LFxWZ32+lRT/8rLDWZGFny41BsUrGCd 2MAPyLrv18WZDMKkG6aJumMp9VravUw4O0ynnHGFytKTdMpN+WaPsCnf6MlKK8zugKKbW/R7 293exhxx+BG2/ye+H3T/z+3HuWng/JOPPzann7LTEbFWbkVhRhvc+f4HzSN9+1idQ7e3Oj39 XPQH35v9roFVkE5Tx+dxYgI5rzA/PWy4N+J9ul85gSvINDtMYYZCc1+njmbg4wMSkoYC8/ig J/zZxc9WrzYnnxjfMQlLID+VW/9AUnEUpF+mQT/kZ5560rS/o01Y1ux1eNt7qFcfX2nWJiyJ btSds/zwmP7m2+asGjX9pDHDCocM4sTND3AOhjz5jB18kMvS+Mh5ov3MWe+Z06udYaPoWflE 92hTGpmRES/ZMEuE6c1JJ+xU/sNkpVkiDkzsMRuOTw3VOf/cPIMYOj9oVN+e9X7gAIOOx+No EEB9bH97W0/ZeNJccmk9A4uEGtVP8x3F0bN6NOQU5Vyk0+FesmSxOfvMaqEDpp9/9pltp/B+ kNlkec+s/3m956V1ianwt79ZLNKmyjtLt3tY13zsccdlBF865czIg4s5kXTKTfkWs9Dy8XjK Nx+wiigqrA7POPVkfzIMSjOsbapU2WmmnUo2pL1EHzWon+2mIZM5uC5tKY6lfcUxJtoe8SZL pkyeZM70HCkGmWXLbx99QXzdp12b2+wAvE4TaXFLjUDOK8xScTBL90CX+yw1mGbj5Q7TY9nC FGYxQZN4mI28pG49E/NM2WbMmG6ub3iNBMVVfFdBRIUe8PggU/u88+1s6SBPAZd1jJgBneN5 8UacZJv+YWJWKmi9GDriv3uewpct+8jcfNMNvtLlxv9g7hxzoeeoRTYonFdedbU1r5437wPT 9MbGfiOilcOhTz/ljbK1trdhPdz1jW8we+65l1m//mfz3LPDTI/uXW0YnF9hZAwm0S4PRBg7 bqK56JK6ZsuWLQY8und9wN6HmQ7ppGHAoknj6824V8fYMDRGY8ZNMFW92eRfftlsZ2Y73/uX IzNtEqgbH9yMdJ94eqg56eST7cxJ0yb/9pVwnVc4rql99ln2eVdeVd90eaCrNRff5s1yT5s2 1XKR9NwZGTSYMluIcsvsPmSLMrmNsMzOID1ZE+Mq7JiNhCXCP446yirSGOSQWWfX+YTNNP9F kgB+l5h9Q73A5v6WqTBHUmyRylQ6HW4x/9Ptqi6Ubpul/dSKsB74vL/zvab/Iw/777oXRjzn WV41M7r9xAAnrGOwpbKsR+dFjtMpp9ybzft0yk35Zo/EKd9oyko+76hzd/4FF9q+/unVqlkv 1XvttbcOjjvW7/KwdlZucD1yY1Kv0XX/tsG6zwodZenSxeaWm260YR8uXWGgu+itd8+HbH+7 hvc53PHemuoLzqtl+7Si9+i4PE5OgArz/9yrowMglQnYXNPsIIUZL/5aNar7Cues9z8wp5xy ahz15cuWmTO9bz5jgzI3Z/5C6zBAd0IQ5t6rlSI9so+4iTb9w0wUzw2DwgalHAoWNiiija6t b0evcI7RLChmesPs51nVTrNKswwyYH2umBAjrYWLl9n1xHIfynW+59AASicaDlEoXR5BnlQf 6vagNUFGWqKgywgaruF578yek2eNNJTpG/99HaJYZfWll8fYmX7d+CAvcxcs8r0RIi6c2FQ7 5SRbPnwGBQoLyicdEMRZuvwTc/gRR+DQ3zDwIgqrWCtouYjCDOVIzB1xs24Yca7rgM6fVtgx 6wzzbnddtB4ACjP3xjO4RZdAWJ2Jbo6Zs+ImkE6HW9rBMCsXXQ/x+cUTq1a1xZR3Ipa8PPLY QPPxihW+4zAo0fACW/Osavb9+NGKVabKwQfbQU9pG5EI2rVJU143Rx19dL7QpVPOfD0gopHT KTflG1FhBmSL8g2AEoFLO3bsML16dDd9e/cMzQ0ccN3aqrVnil3X9+kgkXUbimuwkjzggANt P1viwD8Q/Jh0vOuv7yO7s9HyW8Y9aIsrVqzoL+10lWA9qIm+5TXXNrK6CKwc3biSB+4TE6DC rL5HphURYBNlB8fSOdDKjl7bcGur26xDFMR1N3i9xY8Nm4zQawUxbG2X/nHIfW7a7rn7w3TD g85hLt61ew+r0Eu4zh86Us+9MNKf+ZI42GvlURwP3Nayub8WuM/Dj5qr6te3jYPcB3Ni6zis fAVTpkwZe1k/D4yDZsaxPuSYI3cq9BIH3w0V5wgvvTLGXHb5v+Qx/l47PdENkOaLGVrMgrgb zNExY6vv02WG3Ju3aGVH9mRmEM7bNm3eZBV3OOHCpuWi69CmjRv9TmUNbxQQ390T5VdMIXE/ 1p/c12XnDLvURVzXHVicy4bn1apxhjXn1c56JJz76BMIqzPRzzlzWFwECrvDrWeT3ZkQKbMM Nsu3P29vd6d5qGdv6/0d70EMzna8+x7rpwLr6aA0y8CppJFsn045k6WZDeHplFvec6kMiFC+ xVsLKN/i5Z/s6VA2R774vIEjrUSb2xfV7/JE97lhEya/Zi64sI5/WX7LuACdAEr6xReeZz9X 6ppliz6D9nbuvEVm/wMO8Lxv1+AMs08z/wdUmJXCDHxauZVZUyh3oqRoZUdXXr3G1RXD0qVL TM3qp9vLMjOqFcSwBfs6/XQVZihLu+2609386s9WW2+nOn+i5OprONbKGpTFXn0fNlt+2xIX DR6oMVsga4clj3r9hdyAH3bD6/9tzjmntl3HJkqhhGsewkjCZI9G5xKvcUBDIC9/fIz9ySGD 4xRaia/32nReOgWa77tz5tnGR9+DY5G7Vpix9viEY+NNX9AoNW/R0lzsmZBXrXpSHvN53WDq OoRniBkjjrVZjTZt14qx5Anx8e3wE7wZH3hmlw2K+2bPHP2ejnfZSzK4IJ7AJR730SaQqM5E O+fMXXERSKfDLcs+3AE7KYMeqHQtjTDzAkcyP/74gx1QPemkU+zgoVjnbNm6xS412Wefff0B PHlX6vZclpvIM5Pt0ylnsjSzITydclO+2SDZnXmkfLNDVlg+ha/GfPLxSrsMD5Mq7jby5dHm 8iuutJf1u9yNF3SOSawOHe/JY72o+6zS35ZlL0hH9x+lXyn9P0wcUWEOop36NSrMjsIcZpot SopWdoIqbxB6rQyKKUTQNffeVNPX9+kfJmau3e8wQ9mrf9Xl/rfiwsyYtXt8nX6iY/kBw5wb Aw9YPxG2YcYUzsLEWYzmoZVD936RA/INE/KuD3ZJSWEOYhl0Lex5WmFGnP9MnmgaXVPfje6f Q6HH+m0x19Zy0XUIN2iz8n6PPmYdSsDr98UX1LaDA3rgBo3epRdfaEcU/YclOZAGkwpzElAR C05UZyKWVWYnIgTS6XDL58zcNk6KJIOnMlNx6GGHSVDoXpaEiGWMWGNhNln8T+DmQQMfswN7 MisdmqATkE45nSSy8jSdclO+2SNqyjd7ZKVzCp8M8z6YaydY5DvM6KNinTE+Eavf5bgPyjS+ 5Yy+smwwyYaJ9RF//3ueZYUSJ6jP+sXnn5uqx+9c0iL6hTbHhg6AL9noPEg8SZf71AhQYXYU ZmATUwZBuGjJcvPiCyPsbKpWdoIqr9yj99LhwDWpqFpBlGv6Hhynmr6+T/8owhQljP7XrlXD X3tdwzEFRnpaIURHqt5ll5lNGzfpR8UdYybh4X79je5MoYwow+CBj/vOwfRNUARh9g6zZc0j TGFG43Jnu7bWRBrrqeH1795OHfKYTOtnyHEQy6BrEl/2oqAHdSZhkjh1yhQz/NmhZr7nBM3d 0MGUdeGuXLQDJ20yLsrxypUrfKsEbYGg08HzrvjXVf43pN3ny3n1M88yrVon9nQucbmPDgEt a93uRCeHzEnUCKTT4Za2F+1VkONBcf4Y1AYGlf+77741R//9sECfHW49HjP6ZeskUc/GBKXp XkunnG4a2XieTrkp3+yRNOUbLVnBguaVUS8ZtGmlS5c2be9oH7g0UXKtJ9zQns5buNgcfPAh ccoqBg0XLV1uypYtK7elvA/qs2pfN2KWvXDBfOuwV7fpuj8RpnOknJEcjUiFOUBhRl3Qptkw J97DGyWaO+d938MxHDbpyuuaqun6JB0OXJOOgbzEcC2s8ur0ZfYW8RNt+kfhdk70fXpWE9fd PGgl33VGpdNJ9Xjt2u/MSs8pDL49jG91yibl0jzCzPNQNlmXG2SSLZ88kbT1Hh7H4UwBDcjs 9+dZr4ap8E2kMOv04QDu01Wr7Lo8Wa+OcOGaTC7aZPyDBYvNdM/bdpf77rH5dTuxkic0vB9+ tMKOYOq88LhkEEhWZ0pGKVmKTBJIp8OtZyOeGvqsue76xnFZck37klmqSHw9a6xnmPV65cGD HjedOtyZx8lmXAYCTtIpZ0AyWXcpnXJTvtkjZso3WrLCb6e657T3yy++sE5lZcY4US71UjrR C/S7PNWBx6BnhPVZdf8RjmhHvfSi9ZukJ810HqRfGvQMXgsnQIU5RGHWI0Uan1ZCpROA8CZN bzYDB+ddx4AwUXBwLLOnWkEMq7xhPw6kE7bpH4XOa1B8MZuTML2eWedP/+gkruyxfuKVl18y xx1/vLm1ZWtTsVJF+zmsxYs/NDW8byW3v6ujRPX3iC+u8MXZln6e/gSJf5N3oHmLo7SxY17x nX7BMdnV9f/6jJfcq5mgsVqybKU1BU+Fr8hOGrnynqMyrFdZMH+e2bhpo+85W56FvfaMLvnc tm2bv34kSC4yK4P7sX4FznDQSMv9GKCRTRyR4VwaZAmTPRR4fKf5jOpn2u+MN72lWcKRUbmP ++gQ0PU2qM5EJ6fMSVQIpNPhRt5FycWA4mTPa7V8c37ypAnmumsb2OKlMnAq5oHSXlaqtKe9 d/OmTabGmafbNk2caaJ+i0+KsEFSe3PAv3TLGZBUVl1Kt9yUb3aImfKNlpxg1ShffUHOZMIr LJeI3+KWplZhDZthdtvGsLSCrof1Wb/+6ivfWzaWOk6f/rpddvniqFesFSLS0v2JMJ0j6Jm8 9hcBKswhCjMQuabZuKY7ru5npabNmGlqeObNetNpYEZQPl2kFcSwyhv249Dpu8f6R6Hz6sbD uR55xjniyyeXYIpS96IL/LWyI158yTraQjzZ9GelpHHY2/sWnYzIobx6vZrcp8slP2jNA/GC PrN1a7ObbUOEcPnslOaLhujdOR9Yr6uIIxtmtO9oe5s91QqozofMdMs9sncV5ooVK5mWLZqZ kZ6JPjYZAJH42OvZefFQjTXJ4nAhSC5oaHW6kl5QR1LWpCGOrOUGf73J2kBcEy+1OpzH0SeQ n99y9EvDHBYFgXQ73GjLLzzvHH+ZDtoTePuXDaZ+k1+bZpfPyDV3r5fM6GUkEk8rbd2697TL ft56c4Ztr4PeE3Jf0D7dcgallU3X0i035ZsdUqZ8oycnvTwRuUNf+F9XXm30JAauo8/87NBn 7LJBnKNvhuWccBqs3+WFoTCLWTacL8oaajxHPmOL/Og8hOkciMctnAAV5gQKM7Bp02ycu8rO y6NGmmZNmyDIbkOefMZc+s9/2k8DwXFW4+saSpDRnQitIIZV3lQUOj/x/x3oH4WbVzcuzmFm jplI2bSHaq2YIRxOrK5vfIMpV7ac98H0Jfabm5gFxaaVMukY4bp8o/OYo48x27ZvM2/PfMvc cH0jBFlz48WeSTGcH2geNtD7N3bcRFOtenW7drrrA10MZpOxYZ0vPF3D07araKKRQIN2wokn 2gZs+LCh9tuf9kbvn1bEU+HrKsyYMdHfTsbzHh88xNQ651yzS6ldLBco59JoiWKfilxcR2sY cNDmi1IGdzADeRg1eqw54YQT7eesRgx/1n6sXuIvW/lp3Npyuc59tAmkUmeiXQLmrqgJpNvh Rj6hVDW/+SYDJVZvGPR7qGefhMqy3H9WtdMMvp4A3w3uIB4c4+CrBs89O8xPHm3X1OlvmqOP Oca/lspBQcqZSvpRjVOQclO+UZXqX/mifP9iEZUjUUZh9Scb2ra7O93r9TOrWsUZ/b1+fXvH +eoZ8+oEc8ml9ewt+l2ONm/R0mVGrG8kzVT2ifqs2iwbaenJIZzrPITpHIjHLZxAzirMshY2 WcUJmoXVDpugsD3U7cGEHzQHfnyvt3ffR3wHTVpBDMtDoh9HmEjxo5CyQWEO+p6xe68ohbiO H/Os9+eaKlUOttFgfgwT4EQbFLv3Pphv9t57HxsNzrDOO6emP1sRdi++/Swm25pHWHxcR/7c WeSNGzeYK+rVtRYBie51TbZT4Svmz7qRg8zF7CbR82ooZ2q6HoXJRcdBuuJhNugZ+NTLJXXO T8pYTN6D0uC1aBPQ9SGszkS7BMxdURMoSIdb8rpmzTf2M3X4dMrfKv7NDmhKWCb233+/1rNu +s3stvtu9j3jztSk8oxMlDOV50QtTibKTflGTap/5Yfy/YtFlI7Qx2x6Y2PPt8zrKWULE2dY WiebfpfrvqSEp7pP1GfVZtlIb/TY8XaCT9LWukGYziFxuQ8mkLMKs5jHpqJQaLNfd9RGsL7/ 3mxze5tW/syiXMcMa/eeve0MpFzDXn/bMsjRCuLomUw4Gzj2uONwOeGmR5GwFhjrqpN1SFyv 2dpZCx6G9botW9ySp2wI6/Pwo3b9tuvxD+bqcLSFGXp3A5MevfuammfX8oO0wnz3PffZ2eMe 3bv64TjAoEOX+7t6nbiKcddxgllXrI2G2ba7wUFY5/sfNP846qi4IDR+V//rMntN1tXFRfBO ZLbcNX22z/O8J8IxF8xg9IbRx/6PDbSu/MVBjv5MVCK5DHi0n00T6SWTOWZsnn5yiB9f5+HC OheZLg9289Yvn6Yv8ziLCOC3LGs8E9WZLCoSs1rIBDLR4S7kLGYk+VwppwsrV8qdK+WkfHd+ p9jlENXzd96eaSfIYJkZtMEap3WbO6w5tg7X/T/0Jed6X4cpV768jpLSsdYJ3D4rJnIaXVvf TJk8yU4szf9wiT+JhcS1bpCK3pNShnIsUk4qzIUpY4yew4QDf3BDv99++xfm44o0bZkZKF+h vNm6ZavZ/4ADTJkyZRLmAUodZpxhPg0lE/Fhgu1uWmEW03Xch4YGMx3lvcZlL299dLINcTEY gU9VoYHAd+2w7riwNjzvhx/Wme3btpvdS+9u8ImoypUPsuUtrGe66YLxuu+/N2XLlbUzN3C8 lgorNx2ekwAJZDeBXFE0cqWcbm3MlXLnSjkp3+xSmEVemBDa5DkyxGbXKHv94f322y8tJVjS 5D76BKgwR19GOZFDrTDTXCQnRM5CkgAJZJhArigauVJOt3rkSrlzpZyUb3YqzK7ceJ4bBKgw 54acI19KKsyRFxEzSAIkEHEComhEPJsZy17dy3Krw035ZqzqRDIhyjeSYmGmSMASoMLMihAJ AlSYIyEGZoIESCCLCbDDncXCSyHrlG8KkLI4CuWbxcJj1ks8ASrMJV7E2VFAKszZISfmkgRI gARIgARIgARIgARyiQAV5lySdsTLutlzogBPf2XLlStSp1kRx8LskQAJkAAJkAAJkAAJkAAJ FBMBKszFBJ6PJQESIAESIAESIAESIAESIAESiDYBKszRlg9zRwIkQAIkQAIkQAIkQAIkQAIk UEwEqDAXE3g+lgRIgARIgARIgARIgARIgARIINoEqDBHWz7MHQmQAAmQAAmQAAmQAAmQAAmQ QDERoMJcTOD5WBIgARIgARIgARIgARIgARIggWgToMIcbfkwdyRAAiRAAiRAAiRAAiRAAiRA AsVEgApzMYHnY0mABEiABEiABEiABEiABEiABKJNgApztOXD3JEACZAACZAACZAACZAACZAA CRQTASrMxQSejyUBEiABEiABEiABEiABEiABEog2ASrM0ZYPc0cCJEACJEACJEACJEACJEAC JFBMBP4fx/BRetGKoRkAAAAASUVORK5CYII= --------------627018FE70FE823EEDB7FEB5-- --------------52C86E47EED6E2100703EC63--
--_000_EF8A6CD043B349ADB14F59EB42E5B511acroniscom_ Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 TmlyLA0KDQpJc27igJl0IFNQTSBtYW5hZ2VkIHZpYSBTYW5sb2NrPyBJIGJlbGlldmUgdGhlcmUg aXMgbm8gbmVlZCB0byBmZW5jZSBTUE0gaG9zdC4gRXNwZWNpYWxseSBpZiB0aGVyZSBhcmUgbm8g U1BNIHRhc2tzIHJ1bm5pbmcuDQoNCg0KRnJvbTogPHVzZXJzLWJvdW5jZXNAb3ZpcnQub3JnPiBv biBiZWhhbGYgb2YgTmlyIFNvZmZlciA8bnNvZmZlckByZWRoYXQuY29tPg0KRGF0ZTogTW9uZGF5 LCAxNyBBcHJpbCAyMDE3IGF0IDE2OjA2DQpUbzogS29uc3RhbnRpbiBSYXNrb3NobnlpIDxrb25y YXNrb0BnbWFpbC5jb20+LCBEYW4gWWFzbnkgPGR5YXNueUBnbWFpbC5jb20+DQpDYzogdXNlcnMg PHVzZXJzQG92aXJ0Lm9yZz4sIEZFUk5BTkRPIEZSRURJQU5JIDxmZXJuYW5kby5mcmVkaWFuaUB1 cHguY29tPg0KU3ViamVjdDogUmU6IFtvdmlydC11c2Vyc10gc3RvcmFnZSByZWR1bmRhbmN5IGlu IE92aXJ0DQoNCk9uIE1vbiwgQXByIDE3LCAyMDE3IGF0IDg6MjQgQU0gS29uc3RhbnRpbiBSYXNr b3NobnlpIDxrb25yYXNrb0BnbWFpbC5jb208bWFpbHRvOmtvbnJhc2tvQGdtYWlsLmNvbT4+IHdy b3RlOg0KQnV0IGFjdHVhbGx5LCBpdCBkaWRuJ3Qgd29yayB3ZWxsLiBBZnRlciBtYWluIFNQTSBo b3N0IHdlbnQgZG93biBJIHNlZSB0aGlzDQoNCg0KMjAxNy0wNC0xNyAwNToyMzoxNSw1NTRaIEVS Uk9SIFtvcmcub3ZpcnQuZW5naW5lLmNvcmUudmRzYnJva2VyLmlyc2Jyb2tlci5JcnNQcm94eV0g KERlZmF1bHRRdWFydHpTY2hlZHVsZXI1KSBbNGRjYzAzM2QtMjZiZi00OWJiLWJmYWEtMDNhOTcw ZGJiZWMxXSBTUE0gSW5pdDogY291bGQgbm90IGZpbmQgcmVwb3J0ZWQgdmRzIG9yIG5vdCB1cCAt IHBvb2w6ICdTVEcnIHZkc19zcG1faWQ6ICcxJw0KMjAxNy0wNC0xNyAwNToyMzoxNSw1NjdaIElO Rk8gIFtvcmcub3ZpcnQuZW5naW5lLmNvcmUudmRzYnJva2VyLmlyc2Jyb2tlci5JcnNQcm94eV0g KERlZmF1bHRRdWFydHpTY2hlZHVsZXI1KSBbNGRjYzAzM2QtMjZiZi00OWJiLWJmYWEtMDNhOTcw ZGJiZWMxXSBTUE0gc2VsZWN0aW9uIC0gdmRzIHNlZW1zIGFzIHNwbSAndGFuazUnDQoyMDE3LTA0 LTE3IDA1OjIzOjE1LDU2N1ogV0FSTiAgW29yZy5vdmlydC5lbmdpbmUuY29yZS52ZHNicm9rZXIu aXJzYnJva2VyLklyc1Byb3h5XSAoRGVmYXVsdFF1YXJ0elNjaGVkdWxlcjUpIFs0ZGNjMDMzZC0y NmJmLTQ5YmItYmZhYS0wM2E5NzBkYmJlYzFdIHNwbSB2ZHMgaXMgbm9uIHJlc3BvbnNpdmUsIHN0 b3BwaW5nIHNwbSBzZWxlY3Rpb24uDQoNClNvIHRoYXQgbWVhbnMgb25seSBpZiBCTUMgaXMgdXAg aXQncyBwb3NzaWJsZSB0byBhdXRvbWF0aWNhbGx5IHN3aXRjaCAgU1BNIGhvc3Q/DQoNCkJNQz8N Cg0KSWYgeW91ciBTUE0gaXMgbm8gcmVzcG9uc2l2ZSwgdGhlIHN5c3RlbSB3aWxsIHRyeSB0byBm ZW5jZSBpdC4gRGlkIHlvdQ0KY29uZmlndXJlIHBvd2VyIG1hbmFnZW1lbnQgZm9yIGFsbCBob3N0 cz8gZGlkIHlvdSBjaGVjayB0aGF0IGl0DQp3b3JrPyBIb3cgZGlkIHlvdSBzaW11bGF0ZSBub24t cmVzcG9uc2l2ZSBob3N0Pw0KDQpJZiBwb3dlciBtYW5hZ2VtZW50IGlzIG5vdCBjb25maWd1cmVk IG9yIGZhaWwsIHRoZSBzeXN0ZW0gY2Fubm90DQptb3ZlIHRoZSBzcG0gdG8gYW5vdGhlciBob3N0 LCB1bmxlc3MgeW91IG1hbnVhbGx5IGNvbmZpcm0gdGhhdCB0aGUNClNQTSBob3N0IHdhcyByZWJv b3RlZC4NCg0KTmlyDQoNCg0KVGhhbmtzDQoNCk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDg6Mjkg UE0sIEtvbnN0YW50aW4gUmFza29zaG55aSA8a29ucmFza29AZ21haWwuY29tPG1haWx0bzprb25y YXNrb0BnbWFpbC5jb20+PiB3cm90ZToNCk9oLCBmZW5jZSBhZ2VudCB3b3JrcyBmaW5lIGlmIEkg c2VsZWN0IGlsbzQsDQpUaGFuayB5b3UgZm9yIHlvdXIgaGVscCENCg0KT24gU3VuLCBBcHIgMTYs IDIwMTcgYXQgODoyMiBQTSBEYW4gWWFzbnkgPGR5YXNueUBnbWFpbC5jb208bWFpbHRvOmR5YXNu eUBnbWFpbC5jb20+PiB3cm90ZToNCk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDExOjE5IFBNLCBL b25zdGFudGluIFJhc2tvc2hueWkgPGtvbnJhc2tvQGdtYWlsLmNvbTxtYWlsdG86a29ucmFza29A Z21haWwuY29tPj4gd3JvdGU6DQpNYWtlcyBzZW5zZS4NCkkgd2FzIHRyeWluZyB0byBzZXQgaXQg dXAsIGJ1dCBkb2Vzbid0IHdvcmsgd2l0aCBvdXIgc3RhZ2luZyBoYXJkd2FyZS4NCldlIGhhdmUg b2xkIGlsbzEwMCwgSSdsbCB0cnkgYWdhaW4uDQpUaGFua3MhDQoNCg0KSXQgaXMgYWJzb2x1dGVs eSBuZWNlc3NhcnkgZm9yIGFueSBIQSB0byB3b3JrIHByb3Blcmx5LiBUaGVyZSdzIG9mIGNvdXJz ZSB0aGUgImNvbmZpcm0gaG9zdCBoYXMgYmVlbiBzaHV0ZG93biIgb3B0aW9uLCB3aGljaCBzZXJ2 ZXMgYXMgYW4gb3ZlcnJpZGUgZm9yIHRoZSBmZW5jZSBjb21tYW5kLCBidXQgaXQncyBtYW51YWwN Cg0KT24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgODoxOCBQTSBEYW4gWWFzbnkgPGR5YXNueUBnbWFp bC5jb208bWFpbHRvOmR5YXNueUBnbWFpbC5jb20+PiB3cm90ZToNCk9uIFN1biwgQXByIDE2LCAy MDE3IGF0IDExOjE1IFBNLCBLb25zdGFudGluIFJhc2tvc2hueWkgPGtvbnJhc2tvQGdtYWlsLmNv bTxtYWlsdG86a29ucmFza29AZ21haWwuY29tPj4gd3JvdGU6DQpGZW5jZSBhZ2VudCB1bmRlciBl YWNoIG5vZGU/DQoNCldoZW4geW91IGNvbmZpZ3VyZSBhIGhvc3QsIHRoZXJlJ3MgdGhlIHBvd2Vy IG1hbmFnZW1lbnQgdGFiLCB3aGVyZSB5b3UgbmVlZCB0byBlbnRlciB0aGUgYm1jIGRldGFpbHMg Zm9yIHRoZSBob3N0LiBJZiB5b3UgZG9uJ3QgaGF2ZSBmZW5jaW5nIGVuYWJsZWQsIGhvdyBkbyB5 b3UgZXhwZWN0IHRoZSBzeXN0ZW0gdG8gbWFrZSBzdXJlIGEgaG9zdCBydW5uaW5nIGEgc2Vydmlj ZSBpcyBhY3R1YWxseSBkb3duIChhbmQgaXQgaXMgc2FmZSB0byBzdGFydCBIQSBzZXJ2aWNlcyBl bHNld2hlcmUpLCBhbmQgbm90LCBmb3IgZXhhbXBsZSwganVzdCB1bnJlYWNoYWJsZSBieSB0aGUg ZW5naW5lPyBIb3cgZG8geW91IGF2b2lkIGEgc3BsaXRicmFpZCAtPiBTQkEgPw0KDQoNCk9uIFN1 biwgQXByIDE2LCAyMDE3IGF0IDg6MTQgUE0gRGFuIFlhc255IDxkeWFzbnlAZ21haWwuY29tPG1h aWx0bzpkeWFzbnlAZ21haWwuY29tPj4gd3JvdGU6DQpPbiBTdW4sIEFwciAxNiwgMjAxNyBhdCAx MToxMyBQTSwgS29uc3RhbnRpbiBSYXNrb3NobnlpIDxrb25yYXNrb0BnbWFpbC5jb208bWFpbHRv OmtvbnJhc2tvQGdtYWlsLmNvbT4+IHdyb3RlOg0KIkNvcm5lciBjYXNlcyI/DQpJIHRyaWVkIHRv IHNpbXVsYXRlIGNyYXNoIG9mIFNQTSBzZXJ2ZXIgYW5kIG92aXJ0IGtlcHQgdHJ5aW5nIHRvIHJl aXN0YWJsaXNoZWQgY29ubmVjdGlvbiB0byB0aGUgZmFpbGVkIG5vZGUuDQoNCkRpZCB5b3UgY29u ZmlndXJlIGZlbmNpbmc/DQoNCg0KDQpPbiBTdW4sIEFwciAxNiwgMjAxNyBhdCA4OjEwIFBNIERh biBZYXNueSA8ZHlhc255QGdtYWlsLmNvbTxtYWlsdG86ZHlhc255QGdtYWlsLmNvbT4+IHdyb3Rl Og0KT24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgNzoyOSBBTSwgTmlyIFNvZmZlciA8bnNvZmZlckBy ZWRoYXQuY29tPG1haWx0bzpuc29mZmVyQHJlZGhhdC5jb20+PiB3cm90ZToNCk9uIFN1biwgQXBy IDE2LCAyMDE3IGF0IDI6MDUgUE0gRGFuIFlhc255IDxkeWFzbnlAcmVkaGF0LmNvbTxtYWlsdG86 ZHlhc255QHJlZGhhdC5jb20+PiB3cm90ZToNCg0KDQpPbiBBcHIgMTYsIDIwMTcgNzowMSBBTSwg Ik5pciBTb2ZmZXIiIDxuc29mZmVyQHJlZGhhdC5jb208bWFpbHRvOm5zb2ZmZXJAcmVkaGF0LmNv bT4+IHdyb3RlOg0KT24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgNDoxNyBBTSBEYW4gWWFzbnkgPGR5 YXNueUBnbWFpbC5jb208bWFpbHRvOmR5YXNueUBnbWFpbC5jb20+PiB3cm90ZToNCldoZW4geW91 IHNldCB1cCBhIHN0b3JhZ2UgZG9tYWluLCB5b3UgbmVlZCB0byBzcGVjaWZ5IGEgaG9zdCB0byBw ZXJmb3JtIHRoZSBpbml0aWFsIHN0b3JhZ2Ugb3BlcmF0aW9ucywgYnV0IG9uY2UgdGhlIFNEIGlz IGRlZmluZWQsIGl0J3MgZGV0YWlscyBhcmUgaW4gdGhlIGVuZ2luZSBkYXRhYmFzZSwgYW5kIGFs bCB0aGUgaG9zdHMgZ2V0IGNvbm5lY3RlZCB0byBpdCBkaXJlY3RseS4gSWYgdGhlIGZpcnN0IGhv c3QgeW91IHVzZWQgdG8gZGVmaW5lIHRoZSBTRCBnb2VzIGRvd24sIGFsbCBvdGhlciBob3N0cyB3 aWxsIHN0aWxsIHJlbWFpbiBjb25uZWN0ZWQgYW5kIHdvcmsuIFNQTSBpcyBhbiBIQSBzZXJ2aWNl LCBhbmQgaWYgdGhlIGN1cnJlbnQgU1BNIGhvc3QgZ29lcyBkb3duLCBTUE0gZ2V0cyBzdGFydGVk IG9uIGFub3RoZXIgaG9zdCBpbiB0aGUgREMuIEluIHNob3J0LCB1bmxlc3MgeW91ciBhY3R1YWwg TkZTIGV4cG9ydGluZyBob3N0IGdvZXMgZG93biwgdGhlcmUgaXMgbm8gb3V0YWdlLg0KDQpUaGVy ZSBpcyBubyBzdG9yYWdlIG91dGFnZSwgYnV0IGlmIHlvdSBzaHV0ZG93biB0aGUgc3BtIGhvc3Qs IHRoZSBzcG0gaG9zdA0Kd2lsbCBub3QgbW92ZSB0byBhIG5ldyBob3N0IHVudGlsIHRoZSBzcG0g aG9zdCBpcyBvbmxpbmUgYWdhaW4sIG9yIHlvdSBjb25maXJtDQptYW51YWxseSB0aGF0IHRoZSBz cG0gaG9zdCB3YXMgcmVib290ZWQuDQoNCkluIGEgcHJvcGVybHkgY29uZmlndXJlZCBzZXR1cCB0 aGUgU0JBIHNob3VsZCB0YWtlIGNhcmUgb2YgdGhhdC4gVGhhdCdzIHRoZSB3aG9sZSBwb2ludCBv ZiBIQSBzZXJ2aWNlcw0KDQpJbiBzb21lIGNhc2VzIGxpa2UgcG93ZXIgbG9zcyBvciBoYXJkd2Fy ZSBmYWlsdXJlLCB0aGVyZSBpcyBubyB3YXkgdG8gc3RhcnQNCnRoZSBzcG0gaG9zdCwgYW5kIHRo ZSBzeXN0ZW0gY2Fubm90IHJlY292ZXIgYXV0b21hdGljYWxseS4NCg0KVGhlcmUgYXJlIGFsd2F5 cyBjb3JuZXIgY2FzZXMsIG5vIGRvdWJ0LiBCdXQgaW4gYSBub3JtYWwgc2l0dWF0aW9uLiB3aGVy ZSBhbiBTUE0gaG9zdCBnb2VzIGRvd24gYmVjYXVzZSBvZiBhIGhhcmR3YXJlIGZhaWx1cmUsIGl0 IGdldHMgZmVuY2VkLCBvdGhlciBob3N0cyBjb250ZW5kIGZvciBTUE0gYW5kIHN0YXJ0IGl0LiBO byBzdXJwcmlzZXMgdGhlcmUuDQoNCg0KTmlyDQoNCg0KDQpOaXINCg0KDQpPbiBTYXQsIEFwciAx NSwgMjAxNyBhdCAxOjUzIFBNLCBLb25zdGFudGluIFJhc2tvc2hueWkgPGtvbnJhc2tvQGdtYWls LmNvbTxtYWlsdG86a29ucmFza29AZ21haWwuY29tPj4gd3JvdGU6DQpIaSBGZXJuYW5kbywNCkkg c2VlIGVhY2ggaG9zdCBoYXMgZGlyZWN0IGNvbm5lY3Rpb24gbmZzIG1vdW50LCBidXQgeWVzLCBp ZiBtYWluIGhvc3QgdG8gd2hpY2ggSSBjb25uZWN0ZWQgbmZzIHN0b3JhZ2UgZ29pbmcgZG93biB0 aGUgc3RvcmFnZSBiZWNvbWVzIHVuYXZhaWxhYmxlIGFuZCBhbGwgdm1zIGFyZSBkb3duDQoNCg0K T24gU2F0LCBBcHIgMTUsIDIwMTcgYXQgMTA6MzcgQU0gRkVSTkFORE8gRlJFRElBTkkgPGZlcm5h bmRvLmZyZWRpYW5pQHVweC5jb208bWFpbHRvOmZlcm5hbmRvLmZyZWRpYW5pQHVweC5jb20+PiB3 cm90ZToNCkhlbGxvIEtvbnN0YW50aW4uDQpUaGF0IGRvZXNuYHQgbWFrZSBtdWNoIHNlbnNlIG1h a2UgYSB3aG9sZSBjbHVzdGVyIGRlcGVuZCBvbiBhIHNpbmdsZSBob3N0LiBGcm9tIHdoYXQgSSBr bm93IGFueSBob3N0IHRhbGsgZGlyZWN0bHkgdG8gTkZTIFN0b3JhZ2UgQXJyYXkgb3Igd2hhdGV2 ZXIgb3RoZXIgU2hhcmVkIFN0b3JhZ2UgeW91IGhhdmUuDQpIYXZlIHlvdSB0ZXN0ZWQgdGhhdCBo b3N0IGdvaW5nIGRvd24gaWYgdGhhdCBhZmZlY3RzIHRoZSBvdGhlciB3aXRoIHRoZSBORlMgbW91 bnRlZCBkaXJlY3RsbHkgaW4gYSBORlMgU3RvcmFnZSBhcnJheSA/DQpGZXJuYW5kbw0KDQoyMDE3 LTA0LTE1IDEyOjQyIEdNVC0wMzowMCBLb25zdGFudGluIFJhc2tvc2hueWkgPGtvbnJhc2tvQGdt YWlsLmNvbTxtYWlsdG86a29ucmFza29AZ21haWwuY29tPj46DQpJbiBvdmlydCB5b3UgaGF2ZSB0 byBhdHRhY2ggc3RvcmFnZSB0aHJvdWdoIHNwZWNpZmljIGhvc3QuDQpJZiBob3N0IGdvZXMgZG93 biBzdG9yYWdlIGlzIG5vdCBhdmFpbGFibGUuDQoNCk9uIFNhdCwgQXByIDE1LCAyMDE3IGF0IDc6 MzEgQU0gRkVSTkFORE8gRlJFRElBTkkgPGZlcm5hbmRvLmZyZWRpYW5pQHVweC5jb208bWFpbHRv OmZlcm5hbmRvLmZyZWRpYW5pQHVweC5jb20+PiB3cm90ZToNCldlbGwsIG1ha2UgaXQgbm90IGdv IHRocm91Z2ggaG9zdDEgYW5kIGRlZGljYXRlIGEgc3RvcmFnZSBzZXJ2ZXIgZm9yIHJ1bm5pbmcg TkZTIGFuZCBtYWtlIGJvdGggaG9zdHMgY29ubmVjdCB0byBpdC4NCkluIG15IHZpZXcgTkZTIGlz IG11Y2ggZWFzaWVyIHRvIG1hbmFnZSB0aGFuIGFueSBvdGhlciB0eXBlIG9mIHN0b3JhZ2UsIHNw ZWNpYWxseSBGQyBhbmQgaVNDU0kgYW5kIHBlcmZvcm1hbmNlIGlzIHByZXR0eSBtdWNoIHRoZSBz YW1lLCBzbyB5b3Ugd29uYHQgZ2V0IGJldHRlciByZXN1bHRzIG90aGVyIHRoYW4gbWFuYWdlbWVu dCBnb2luZyB0byBvdGhlciB0eXBlLg0KRmVybmFuZG8NCg0KMjAxNy0wNC0xNSA1OjI1IEdNVC0w MzowMCBLb25zdGFudGluIFJhc2tvc2hueWkgPGtvbnJhc2tvQGdtYWlsLmNvbTxtYWlsdG86a29u cmFza29AZ21haWwuY29tPj46DQpIaSBndXlzLA0KSSBoYXZlIG9uZSBuZnMgc3RvcmFnZSwNCml0 J3MgY29ubmVjdGVkIHRocm91Z2ggaG9zdDEuDQpob3N0MiBhbHNvIGhhcyBhY2Nlc3MgdG8gaXQs IEkgY2FuIGVhc2lseSBtaWdyYXRlIHZtcyBiZXR3ZWVuIHRoZW0uDQoNClRoZSBxdWVzdGlvbiBp cyAtIGlmIGhvc3QxIGlzIGRvd24gLSBhbGwgaW5mcmFzdHJ1Y3R1cmUgaXMgZG93biwgc2luY2Ug YWxsIHRyYWZmaWMgZ29lcyB0aHJvdWdoIGhvc3QxLA0KaXMgdGhlcmUgYW55IHdheSBpbiBvVmly dCB0byB1c2UgcmVkdW5kYW50IHN0b3JhZ2U/DQoNCk9ubHkgZ2x1c3RlcmZzPw0KDQpUaGFua3MN Cg0KDQpfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXw0KVXNl cnMgbWFpbGluZyBsaXN0DQpVc2Vyc0BvdmlydC5vcmc8bWFpbHRvOlVzZXJzQG92aXJ0Lm9yZz4N Cmh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0aW5mby91c2Vycw0KDQoNCl9fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQpVc2VycyBtYWlsaW5n IGxpc3QNClVzZXJzQG92aXJ0Lm9yZzxtYWlsdG86VXNlcnNAb3ZpcnQub3JnPg0KaHR0cDovL2xp c3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL3VzZXJzDQoNCl9fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQpVc2VycyBtYWlsaW5nIGxpc3QNClVzZXJz QG92aXJ0Lm9yZzxtYWlsdG86VXNlcnNAb3ZpcnQub3JnPg0KaHR0cDovL2xpc3RzLm92aXJ0Lm9y Zy9tYWlsbWFuL2xpc3RpbmZvL3VzZXJzDQoNCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fDQpVc2VycyBtYWlsaW5nIGxpc3QNClVzZXJzQG92aXJ0Lm9yZzxt YWlsdG86VXNlcnNAb3ZpcnQub3JnPg0KaHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xp c3RpbmZvL3VzZXJzDQoNCg0K --_000_EF8A6CD043B349ADB14F59EB42E5B511acroniscom_ Content-Type: text/html; charset="utf-8" Content-ID: <351727B7CD9E1A4AA911DECC8CBB0E8A@acronis.com> Content-Transfer-Encoding: base64 PGh0bWwgeG1sbnM6bz0idXJuOnNjaGVtYXMtbWljcm9zb2Z0LWNvbTpvZmZpY2U6b2ZmaWNlIiB4 bWxuczp3PSJ1cm46c2NoZW1hcy1taWNyb3NvZnQtY29tOm9mZmljZTp3b3JkIiB4bWxuczptPSJo dHRwOi8vc2NoZW1hcy5taWNyb3NvZnQuY29tL29mZmljZS8yMDA0LzEyL29tbWwiIHhtbG5zPSJo dHRwOi8vd3d3LnczLm9yZy9UUi9SRUMtaHRtbDQwIj4NCjxoZWFkPg0KPG1ldGEgaHR0cC1lcXVp dj0iQ29udGVudC1UeXBlIiBjb250ZW50PSJ0ZXh0L2h0bWw7IGNoYXJzZXQ9dXRmLTgiPg0KPG1l dGEgbmFtZT0iVGl0bGUiIGNvbnRlbnQ9IiI+DQo8bWV0YSBuYW1lPSJLZXl3b3JkcyIgY29udGVu dD0iIj4NCjxtZXRhIG5hbWU9IkdlbmVyYXRvciIgY29udGVudD0iTWljcm9zb2Z0IFdvcmQgMTUg KGZpbHRlcmVkIG1lZGl1bSkiPg0KPHN0eWxlPjwhLS0NCi8qIEZvbnQgRGVmaW5pdGlvbnMgKi8N CkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6IkNhbWJyaWEgTWF0aCI7DQoJcGFub3NlLTE6MiA0 IDUgMyA1IDQgNiAzIDIgNDt9DQpAZm9udC1mYWNlDQoJe2ZvbnQtZmFtaWx5OkNhbGlicmk7DQoJ cGFub3NlLTE6MiAxNSA1IDIgMiAyIDQgMyAyIDQ7fQ0KLyogU3R5bGUgRGVmaW5pdGlvbnMgKi8N CnAuTXNvTm9ybWFsLCBsaS5Nc29Ob3JtYWwsIGRpdi5Nc29Ob3JtYWwNCgl7bWFyZ2luOjBjbTsN CgltYXJnaW4tYm90dG9tOi4wMDAxcHQ7DQoJZm9udC1zaXplOjEyLjBwdDsNCglmb250LWZhbWls eToiVGltZXMgTmV3IFJvbWFuIjt9DQphOmxpbmssIHNwYW4uTXNvSHlwZXJsaW5rDQoJe21zby1z dHlsZS1wcmlvcml0eTo5OTsNCgljb2xvcjpibHVlOw0KCXRleHQtZGVjb3JhdGlvbjp1bmRlcmxp bmU7fQ0KYTp2aXNpdGVkLCBzcGFuLk1zb0h5cGVybGlua0ZvbGxvd2VkDQoJe21zby1zdHlsZS1w cmlvcml0eTo5OTsNCgljb2xvcjpwdXJwbGU7DQoJdGV4dC1kZWNvcmF0aW9uOnVuZGVybGluZTt9 DQpzcGFuLm0tNzA0OTEzMTYxOTY3MzQ1NzA3N20tMzQ0NzQ3MzU3Nzk3OTk3MDUxM20xNzc0MTA4 MDMxOTMyNTkyODg2bTM4NDY1MjU1MTA3OTE2MTY1MjhtMTEyMDM1NzYwODA1NjY4NDc0M20zNjIz NzUwMzA3MTY0OTY2NjJtLTc5NTQxMzA3MjIyMzU0MTUyMTBtMjE3MzkyNTI0MTU3NzMzNjkzNW0t NjQzNDg0Njg3ODMxOTM3MTE5MWhvZW56Yg0KCXttc28tc3R5bGUtbmFtZTptXy03MDQ5MTMxNjE5 NjczNDU3MDc3bV8tMzQ0NzQ3MzU3Nzk3OTk3MDUxM21fMTc3NDEwODAzMTkzMjU5Mjg4Nm1fMzg0 NjUyNTUxMDc5MTYxNjUyOG1fMTEyMDM1NzYwODA1NjY4NDc0M21fMzYyMzc1MDMwNzE2NDk2NjYy bV8tNzk1NDEzMDcyMjIzNTQxNTIxMG1fMjE3MzkyNTI0MTU3NzMzNjkzNW1fLTY0MzQ4NDY4Nzgz MTkzNzExOTFob2VuemI7fQ0Kc3Bhbi5FbWFpbFN0eWxlMTgNCgl7bXNvLXN0eWxlLXR5cGU6cGVy c29uYWwtcmVwbHk7DQoJZm9udC1mYW1pbHk6Q2FsaWJyaTsNCgljb2xvcjp3aW5kb3d0ZXh0O30N CnNwYW4ubXNvSW5zDQoJe21zby1zdHlsZS10eXBlOmV4cG9ydC1vbmx5Ow0KCW1zby1zdHlsZS1u YW1lOiIiOw0KCXRleHQtZGVjb3JhdGlvbjp1bmRlcmxpbmU7DQoJY29sb3I6dGVhbDt9DQouTXNv Q2hwRGVmYXVsdA0KCXttc28tc3R5bGUtdHlwZTpleHBvcnQtb25seTsNCglmb250LXNpemU6MTAu MHB0O30NCkBwYWdlIFdvcmRTZWN0aW9uMQ0KCXtzaXplOjU5NS4wcHQgODQyLjBwdDsNCgltYXJn aW46Mi4wY20gNDIuNXB0IDIuMGNtIDMuMGNtO30NCmRpdi5Xb3JkU2VjdGlvbjENCgl7cGFnZTpX b3JkU2VjdGlvbjE7fQ0KLS0+PC9zdHlsZT4NCjwvaGVhZD4NCjxib2R5IGJnY29sb3I9IndoaXRl IiBsYW5nPSJFTi1HQiIgbGluaz0iYmx1ZSIgdmxpbms9InB1cnBsZSI+DQo8ZGl2IGNsYXNzPSJX b3JkU2VjdGlvbjEiPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6 ZToxMS4wcHQ7Zm9udC1mYW1pbHk6Q2FsaWJyaTttc28tZmFyZWFzdC1sYW5ndWFnZTpFTi1VUyI+ TmlyLDxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0 eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OkNhbGlicmk7bXNvLWZhcmVhc3QtbGFu Z3VhZ2U6RU4tVVMiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29O b3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OkNhbGlicmk7 bXNvLWZhcmVhc3QtbGFuZ3VhZ2U6RU4tVVMiPklzbuKAmXQgU1BNIG1hbmFnZWQgdmlhIFNhbmxv Y2s/IEkgYmVsaWV2ZSB0aGVyZSBpcyBubyBuZWVkIHRvIGZlbmNlIFNQTSBob3N0LiBFc3BlY2lh bGx5IGlmIHRoZXJlIGFyZSBubyBTUE0gdGFza3MgcnVubmluZy48bzpwPjwvbzpwPjwvc3Bhbj48 L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtm b250LWZhbWlseTpDYWxpYnJpO21zby1mYXJlYXN0LWxhbmd1YWdlOkVOLVVTIj48bzpwPiZuYnNw OzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9u dC1zaXplOjExLjBwdDtmb250LWZhbWlseTpDYWxpYnJpO21zby1mYXJlYXN0LWxhbmd1YWdlOkVO LVVTIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8ZGl2IHN0eWxlPSJib3JkZXI6bm9u ZTtib3JkZXItdG9wOnNvbGlkICNCNUM0REYgMS4wcHQ7cGFkZGluZzozLjBwdCAwY20gMGNtIDBj bSI+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48Yj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6Q2Fs aWJyaTtjb2xvcjpibGFjayI+RnJvbTogPC9zcGFuPg0KPC9iPjxzcGFuIHN0eWxlPSJmb250LWZh bWlseTpDYWxpYnJpO2NvbG9yOmJsYWNrIj4mbHQ7dXNlcnMtYm91bmNlc0BvdmlydC5vcmcmZ3Q7 IG9uIGJlaGFsZiBvZiBOaXIgU29mZmVyICZsdDtuc29mZmVyQHJlZGhhdC5jb20mZ3Q7PGJyPg0K PGI+RGF0ZTogPC9iPk1vbmRheSwgMTcgQXByaWwgMjAxNyBhdCAxNjowNjxicj4NCjxiPlRvOiA8 L2I+S29uc3RhbnRpbiBSYXNrb3NobnlpICZsdDtrb25yYXNrb0BnbWFpbC5jb20mZ3Q7LCBEYW4g WWFzbnkgJmx0O2R5YXNueUBnbWFpbC5jb20mZ3Q7PGJyPg0KPGI+Q2M6IDwvYj51c2VycyAmbHQ7 dXNlcnNAb3ZpcnQub3JnJmd0OywgRkVSTkFORE8gRlJFRElBTkkgJmx0O2Zlcm5hbmRvLmZyZWRp YW5pQHVweC5jb20mZ3Q7PGJyPg0KPGI+U3ViamVjdDogPC9iPlJlOiBbb3ZpcnQtdXNlcnNdIHN0 b3JhZ2UgcmVkdW5kYW5jeSBpbiBPdmlydDxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+ DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5PbiBNb24sIEFwciAx NywgMjAxNyBhdCA4OjI0IEFNIEtvbnN0YW50aW4gUmFza29zaG55aSAmbHQ7PGEgaHJlZj0ibWFp bHRvOmtvbnJhc2tvQGdtYWlsLmNvbSI+a29ucmFza29AZ21haWwuY29tPC9hPiZndDsgd3JvdGU6 PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTti b3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7 bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1z b05vcm1hbCI+QnV0IGFjdHVhbGx5LCBpdCBkaWRuJ3Qgd29yayB3ZWxsLiBBZnRlciBtYWluIFNQ TSBob3N0IHdlbnQgZG93biBJIHNlZSB0aGlzDQo8bzpwPjwvbzpwPjwvcD4NCjxkaXY+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxw IGNsYXNzPSJNc29Ob3JtYWwiPjxicj4NCjIwMTctMDQtMTcgMDU6MjM6MTUsNTU0WiBFUlJPUiBb b3JnLm92aXJ0LmVuZ2luZS5jb3JlLnZkc2Jyb2tlci5pcnNicm9rZXIuSXJzUHJveHldIChEZWZh dWx0UXVhcnR6U2NoZWR1bGVyNSkgWzRkY2MwMzNkLTI2YmYtNDliYi1iZmFhLTAzYTk3MGRiYmVj MV0gU1BNIEluaXQ6IGNvdWxkIG5vdCBmaW5kIHJlcG9ydGVkIHZkcyBvciBub3QgdXAgLSBwb29s OiAnU1RHJyB2ZHNfc3BtX2lkOiAnMSc8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxw IGNsYXNzPSJNc29Ob3JtYWwiPjIwMTctMDQtMTcgMDU6MjM6MTUsNTY3WiBJTkZPICZuYnNwO1tv cmcub3ZpcnQuZW5naW5lLmNvcmUudmRzYnJva2VyLmlyc2Jyb2tlci5JcnNQcm94eV0gKERlZmF1 bHRRdWFydHpTY2hlZHVsZXI1KSBbNGRjYzAzM2QtMjZiZi00OWJiLWJmYWEtMDNhOTcwZGJiZWMx XSBTUE0gc2VsZWN0aW9uIC0gdmRzIHNlZW1zIGFzIHNwbSAndGFuazUnPG86cD48L286cD48L3A+ DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj4yMDE3LTA0LTE3IDA1OjIzOjE1 LDU2N1ogV0FSTiAmbmJzcDtbb3JnLm92aXJ0LmVuZ2luZS5jb3JlLnZkc2Jyb2tlci5pcnNicm9r ZXIuSXJzUHJveHldIChEZWZhdWx0UXVhcnR6U2NoZWR1bGVyNSkgWzRkY2MwMzNkLTI2YmYtNDli Yi1iZmFhLTAzYTk3MGRiYmVjMV0gc3BtIHZkcyBpcyBub24gcmVzcG9uc2l2ZSwgc3RvcHBpbmcg c3BtIHNlbGVjdGlvbi48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCI+U28gdGhhdCBtZWFucyBvbmx5IGlmIEJNQyBpcyB1cCBpdCdzIHBvc3NpYmxl IHRvIGF1dG9tYXRpY2FsbHkgc3dpdGNoICZuYnNwO1NQTSBob3N0PzxvOnA+PC9vOnA+PC9wPg0K PC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs Ij48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiPkJNQz88bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiPjxvOnA+Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+SWYgeW91ciBTUE0gaXMgbm8gcmVzcG9uc2l2ZSwgdGhlIHN5c3RlbSB3aWxsIHRyeSB0 byBmZW5jZSBpdC4gRGlkIHlvdSZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0K PHAgY2xhc3M9Ik1zb05vcm1hbCI+Y29uZmlndXJlIHBvd2VyIG1hbmFnZW1lbnQgZm9yIGFsbCBo b3N0cz8gZGlkIHlvdSBjaGVjayB0aGF0IGl0Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4N CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj53b3JrPyBIb3cgZGlkIHlvdSBzaW11bGF0ZSBu b24tcmVzcG9uc2l2ZSBob3N0PyZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0K PHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+ DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5JZiBwb3dlciBtYW5hZ2VtZW50IGlzIG5vdCBjb25maWd1 cmVkIG9yIGZhaWwsIHRoZSBzeXN0ZW0gY2Fubm90PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxk aXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5tb3ZlIHRoZSBzcG0gdG8gYW5vdGhlciBob3N0LCB1 bmxlc3MgeW91IG1hbnVhbGx5IGNvbmZpcm0gdGhhdCB0aGU8bzpwPjwvbzpwPjwvcD4NCjwvZGl2 Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPlNQTSBob3N0IHdhcyByZWJvb3RlZC48bzpw PjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5i c3A7PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+TmlyPG86 cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj4mbmJzcDs8 bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2Jv cmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDtt YXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tcmlnaHQ6MGNtIj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xh c3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIj5UaGFua3M8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8 ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8ZGl2Pg0K PHAgY2xhc3M9Ik1zb05vcm1hbCI+T24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgODoyOSBQTSwgS29u c3RhbnRpbiBSYXNrb3NobnlpICZsdDs8YSBocmVmPSJtYWlsdG86a29ucmFza29AZ21haWwuY29t IiB0YXJnZXQ9Il9ibGFuayI+a29ucmFza29AZ21haWwuY29tPC9hPiZndDsgd3JvdGU6PG86cD48 L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29s aWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQu OHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPk9oLCBm ZW5jZSBhZ2VudCB3b3JrcyBmaW5lIGlmIEkgc2VsZWN0IGlsbzQsPG86cD48L286cD48L3A+DQo8 L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5UaGFuayB5b3UgZm9yIHlvdXIgaGVs cCE8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNz PSJNc29Ob3JtYWwiPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDg6MjIgUE0gRGFuIFlhc255ICZs dDs8YSBocmVmPSJtYWlsdG86ZHlhc255QGdtYWlsLmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmR5YXNu eUBnbWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGJsb2Nr cXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7 cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tcmlnaHQ6 MGNtIj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPk9uIFN1biwg QXByIDE2LCAyMDE3IGF0IDExOjE5IFBNLCBLb25zdGFudGluIFJhc2tvc2hueWkgJmx0OzxhIGhy ZWY9Im1haWx0bzprb25yYXNrb0BnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5rb25yYXNrb0Bn bWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjxibG9ja3F1b3RlIHN0eWxl PSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNt IDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+TWFrZXMgc2Vuc2UuJm5ic3A7PG86cD48L286cD48L3A+ DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5JIHdhcyB0cnlpbmcgdG8gc2V0 IGl0IHVwLCBidXQgZG9lc24ndCB3b3JrIHdpdGggb3VyIHN0YWdpbmcgaGFyZHdhcmUuJm5ic3A7 PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5XZSBo YXZlIG9sZCBpbG8xMDAsIEknbGwgdHJ5IGFnYWluLiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9k aXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+VGhhbmtzISA8bzpwPjwvbzpwPjwvcD4N CjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+ DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+ DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs Ij5JdCBpcyBhYnNvbHV0ZWx5IG5lY2Vzc2FyeSBmb3IgYW55IEhBIHRvIHdvcmsgcHJvcGVybHku IFRoZXJlJ3Mgb2YgY291cnNlIHRoZSAmcXVvdDtjb25maXJtIGhvc3QgaGFzIGJlZW4gc2h1dGRv d24mcXVvdDsgb3B0aW9uLCB3aGljaCBzZXJ2ZXMgYXMgYW4gb3ZlcnJpZGUgZm9yIHRoZSBmZW5j ZSBjb21tYW5kLCBidXQgaXQncyBtYW51YWw8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+ DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHls ZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBj bSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRp dj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5PbiBT dW4sIEFwciAxNiwgMjAxNyBhdCA4OjE4IFBNIERhbiBZYXNueSAmbHQ7PGEgaHJlZj0ibWFpbHRv OmR5YXNueUBnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5keWFzbnlAZ21haWwuY29tPC9hPiZn dDsgd3JvdGU6PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3Jk ZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAw Y20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8ZGl2Pg0KPGRp dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5PbiBTdW4sIEFwciAxNiwgMjAxNyBhdCAx MToxNSBQTSwgS29uc3RhbnRpbiBSYXNrb3NobnlpICZsdDs8YSBocmVmPSJtYWlsdG86a29ucmFz a29AZ21haWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+a29ucmFza29AZ21haWwuY29tPC9hPiZndDsg d3JvdGU6PG86cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9y ZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21h cmdpbi1sZWZ0OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29O b3JtYWwiPkZlbmNlIGFnZW50IHVuZGVyIGVhY2ggbm9kZT88bzpwPjwvbzpwPjwvcD4NCjwvZGl2 Pg0KPC9ibG9ja3F1b3RlPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7 PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4N CjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+V2hlbiB5b3UgY29uZmlndXJlIGEg aG9zdCwgdGhlcmUncyB0aGUgcG93ZXIgbWFuYWdlbWVudCB0YWIsIHdoZXJlIHlvdSBuZWVkIHRv IGVudGVyIHRoZSBibWMgZGV0YWlscyBmb3IgdGhlIGhvc3QuIElmIHlvdSBkb24ndCBoYXZlIGZl bmNpbmcgZW5hYmxlZCwgaG93IGRvIHlvdSBleHBlY3QgdGhlIHN5c3RlbSB0byBtYWtlIHN1cmUg YSBob3N0IHJ1bm5pbmcgYSBzZXJ2aWNlIGlzIGFjdHVhbGx5IGRvd24gKGFuZA0KIGl0IGlzIHNh ZmUgdG8gc3RhcnQgSEEgc2VydmljZXMgZWxzZXdoZXJlKSwgYW5kIG5vdCwgZm9yIGV4YW1wbGUs IGp1c3QgdW5yZWFjaGFibGUgYnkgdGhlIGVuZ2luZT8gSG93IGRvIHlvdSBhdm9pZCBhIHNwbGl0 YnJhaWQgLSZndDsgU0JBID88bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4N CjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi PiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVy Om5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNt IDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+ DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8ZGl2 Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDg6 MTQgUE0gRGFuIFlhc255ICZsdDs8YSBocmVmPSJtYWlsdG86ZHlhc255QGdtYWlsLmNvbSIgdGFy Z2V0PSJfYmxhbmsiPmR5YXNueUBnbWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNv bGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0 LjhwdDttYXJnaW4tcmlnaHQ6MGNtIj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDExOjEzIFBNLCBLb25zdGFudGluIFJh c2tvc2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzprb25yYXNrb0BnbWFpbC5jb20iIHRhcmdldD0i X2JsYW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4N CjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0ND IDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2lu LXJpZ2h0OjBjbSI+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+JnF1b3Q7Q29ybmVyIGNh c2VzJnF1b3Q7PyZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCI+SSB0cmllZCB0byBzaW11bGF0ZSBjcmFzaCBvZiBTUE0gc2VydmVyIGFuZCBv dmlydCBrZXB0IHRyeWluZyB0byByZWlzdGFibGlzaGVkIGNvbm5lY3Rpb24gdG8gdGhlIGZhaWxl ZCBub2RlLiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjwv ZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFz cz0iTXNvTm9ybWFsIj5EaWQgeW91IGNvbmZpZ3VyZSBmZW5jaW5nPzxvOnA+PC9vOnA+PC9wPg0K PC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxi bG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEu MHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJp Z2h0OjBjbSI+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpw PiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxv OnA+Jm5ic3A7PC9vOnA+PC9wPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5P biBTdW4sIEFwciAxNiwgMjAxNyBhdCA4OjEwIFBNIERhbiBZYXNueSAmbHQ7PGEgaHJlZj0ibWFp bHRvOmR5YXNueUBnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5keWFzbnlAZ21haWwuY29tPC9h PiZndDsgd3JvdGU6PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJi b3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBj bSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8ZGl2Pg0K PGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5PbiBTdW4sIEFwciAxNiwgMjAxNyBh dCA3OjI5IEFNLCBOaXIgU29mZmVyICZsdDs8YSBocmVmPSJtYWlsdG86bnNvZmZlckByZWRoYXQu Y29tIiB0YXJnZXQ9Il9ibGFuayI+bnNvZmZlckByZWRoYXQuY29tPC9hPiZndDsgd3JvdGU6PG86 cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6 c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0 OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCI+T24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgMjowNSBQTSBEYW4gWWFzbnkgJmx0 OzxhIGhyZWY9Im1haWx0bzpkeWFzbnlAcmVkaGF0LmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmR5YXNu eUByZWRoYXQuY29tPC9hPiZndDsgd3JvdGU6PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9j a3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0 O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0 OjBjbSI+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9v OnA+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+PC9w Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPk9uIEFwciAxNiwgMjAxNyA3OjAxIEFNLCAm cXVvdDtOaXIgU29mZmVyJnF1b3Q7ICZsdDs8YSBocmVmPSJtYWlsdG86bnNvZmZlckByZWRoYXQu Y29tIiB0YXJnZXQ9Il9ibGFuayI+bnNvZmZlckByZWRoYXQuY29tPC9hPiZndDsgd3JvdGU6PG86 cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6 c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0 OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxw IGNsYXNzPSJNc29Ob3JtYWwiPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDQ6MTcgQU0gRGFuIFlh c255ICZsdDs8YSBocmVmPSJtYWlsdG86ZHlhc255QGdtYWlsLmNvbSIgdGFyZ2V0PSJfYmxhbmsi PmR5YXNueUBnbWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0K PGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0Mg MS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4t cmlnaHQ6MGNtIj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5XaGVuIHlvdSBzZXQgdXAg YSBzdG9yYWdlIGRvbWFpbiwgeW91IG5lZWQgdG8gc3BlY2lmeSBhIGhvc3QgdG8gcGVyZm9ybSB0 aGUgaW5pdGlhbCBzdG9yYWdlIG9wZXJhdGlvbnMsIGJ1dCBvbmNlIHRoZSBTRCBpcyBkZWZpbmVk LCBpdCdzIGRldGFpbHMgYXJlIGluIHRoZSBlbmdpbmUgZGF0YWJhc2UsIGFuZCBhbGwgdGhlIGhv c3RzIGdldCBjb25uZWN0ZWQgdG8gaXQgZGlyZWN0bHkuIElmIHRoZSBmaXJzdCBob3N0DQogeW91 IHVzZWQgdG8gZGVmaW5lIHRoZSBTRCBnb2VzIGRvd24sIGFsbCBvdGhlciBob3N0cyB3aWxsIHN0 aWxsIHJlbWFpbiBjb25uZWN0ZWQgYW5kIHdvcmsuIFNQTSBpcyBhbiBIQSBzZXJ2aWNlLCBhbmQg aWYgdGhlIGN1cnJlbnQgU1BNIGhvc3QgZ29lcyBkb3duLCBTUE0gZ2V0cyBzdGFydGVkIG9uIGFu b3RoZXIgaG9zdCBpbiB0aGUgREMuIEluIHNob3J0LCB1bmxlc3MgeW91ciBhY3R1YWwgTkZTIGV4 cG9ydGluZyBob3N0IGdvZXMgZG93biwgdGhlcmUNCiBpcyBubyBvdXRhZ2UuJm5ic3A7PG86cD48 L286cD48L3A+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y bWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xh c3M9Ik1zb05vcm1hbCI+VGhlcmUgaXMgbm8gc3RvcmFnZSBvdXRhZ2UsIGJ1dCBpZiB5b3Ugc2h1 dGRvd24gdGhlIHNwbSBob3N0LCB0aGUgc3BtIGhvc3Q8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPndpbGwgbm90IG1vdmUgdG8gYSBuZXcgaG9zdCB1 bnRpbCB0aGUgc3BtIGhvc3QgaXMgb25saW5lIGFnYWluLCBvciB5b3UgY29uZmlybTxvOnA+PC9v OnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+bWFudWFsbHkgdGhh dCB0aGUgc3BtIGhvc3Qgd2FzIHJlYm9vdGVkLjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rp dj4NCjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rp dj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+SW4gYSBwcm9wZXJseSBjb25m aWd1cmVkIHNldHVwIHRoZSBTQkEgc2hvdWxkIHRha2UgY2FyZSBvZiB0aGF0LiBUaGF0J3MgdGhl IHdob2xlIHBvaW50IG9mIEhBIHNlcnZpY2VzJm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4N CjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+ Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+SW4g c29tZSBjYXNlcyBsaWtlIHBvd2VyIGxvc3Mgb3IgaGFyZHdhcmUgZmFpbHVyZSwgdGhlcmUgaXMg bm8gd2F5IHRvIHN0YXJ0PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0i TXNvTm9ybWFsIj50aGUgc3BtIGhvc3QsIGFuZCB0aGUgc3lzdGVtIGNhbm5vdCByZWNvdmVyIGF1 dG9tYXRpY2FsbHkuPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Js b2NrcXVvdGU+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48 L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4N CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5UaGVyZSBhcmUgYWx3YXlzIGNvcm5lciBjYXNl cywgbm8gZG91YnQuIEJ1dCBpbiBhIG5vcm1hbCBzaXR1YXRpb24uIHdoZXJlIGFuIFNQTSBob3N0 IGdvZXMgZG93biBiZWNhdXNlIG9mIGEgaGFyZHdhcmUgZmFpbHVyZSwgaXQgZ2V0cyBmZW5jZWQs IG90aGVyIGhvc3RzIGNvbnRlbmQgZm9yIFNQTSBhbmQgc3RhcnQgaXQuIE5vIHN1cnByaXNlcyB0 aGVyZS4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2 Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPiZuYnNw OzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7 Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0 O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImNvbG9yOiM4ODg4ODgiPjxvOnA+ Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiPjxzcGFuIHN0eWxlPSJjb2xvcjojODg4ODg4Ij5OaXI8bzpwPjwvbzpwPjwvc3Bhbj48L3A+ DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPiZuYnNw OzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7 Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0 O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxk aXY+DQo8ZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNv bGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0 LjhwdDttYXJnaW4tcmlnaHQ6MGNtIj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJjb2xvcjojODg4ODg4Ij48bzpwPiZuYnNwOzwvbzpwPjwv c3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHls ZT0iY29sb3I6Izg4ODg4OCI+TmlyPG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2 Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9k aXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0ND Q0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21h cmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7 PC9vOnA+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPk9uIFNhdCwgQXByIDE1LCAy MDE3IGF0IDE6NTMgUE0sIEtvbnN0YW50aW4gUmFza29zaG55aSAmbHQ7PGEgaHJlZj0ibWFpbHRv OmtvbnJhc2tvQGdtYWlsLmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmtvbnJhc2tvQGdtYWlsLmNvbTwv YT4mZ3Q7IHdyb3RlOjxvOnA+PC9vOnA+PC9wPg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpu b25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2 LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tcmlnaHQ6MGNtIj4NCjxkaXY+DQo8cCBjbGFz cz0iTXNvTm9ybWFsIj5IaSBGZXJuYW5kbyw8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiPkkgc2VlIGVhY2ggaG9zdCBoYXMgZGlyZWN0IGNvbm5lY3Rp b24gbmZzIG1vdW50LCBidXQgeWVzLCBpZiBtYWluIGhvc3QgdG8gd2hpY2ggSSBjb25uZWN0ZWQg bmZzIHN0b3JhZ2UgZ29pbmcgZG93biB0aGUgc3RvcmFnZSBiZWNvbWVzIHVuYXZhaWxhYmxlIGFu ZCBhbGwgdm1zIGFyZSBkb3duPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+ DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8ZGl2 Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPk9uIFNhdCwgQXByIDE1LCAyMDE3IGF0IDEw OjM3IEFNIEZFUk5BTkRPIEZSRURJQU5JICZsdDs8YSBocmVmPSJtYWlsdG86ZmVybmFuZG8uZnJl ZGlhbmlAdXB4LmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmZlcm5hbmRvLmZyZWRpYW5pQHVweC5jb208 L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9 ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20g MGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tcmlnaHQ6MGNtIj4NCjxkaXY+ DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibWFyZ2lu LWJvdHRvbToxMi4wcHQiPkhlbGxvIEtvbnN0YW50aW4uPG86cD48L286cD48L3A+DQo8L2Rpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiPlRoYXQgZG9lc25gdCBtYWtlIG11Y2ggc2Vuc2UgbWFrZSBh IHdob2xlIGNsdXN0ZXIgZGVwZW5kIG9uIGEgc2luZ2xlIGhvc3QuIEZyb20gd2hhdCBJIGtub3cg YW55IGhvc3QgdGFsayBkaXJlY3RseSB0byBORlMgU3RvcmFnZSBBcnJheSBvciB3aGF0ZXZlciBv dGhlciBTaGFyZWQgU3RvcmFnZSB5b3UgaGF2ZS48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPHAg Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1hcmdpbi1ib3R0b206MTIuMHB0Ij5IYXZlIHlvdSB0 ZXN0ZWQgdGhhdCBob3N0IGdvaW5nIGRvd24gaWYgdGhhdCBhZmZlY3RzIHRoZSBvdGhlciB3aXRo IHRoZSBORlMgbW91bnRlZCBkaXJlY3RsbHkgaW4gYSBORlMgU3RvcmFnZSBhcnJheSA/PG86cD48 L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPkZl cm5hbmRvPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs Ij48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj4yMDE3 LTA0LTE1IDEyOjQyIEdNVC0wMzowMCBLb25zdGFudGluIFJhc2tvc2hueWkgJmx0OzxhIGhyZWY9 Im1haWx0bzprb25yYXNrb0BnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5rb25yYXNrb0BnbWFp bC5jb208L2E+Jmd0Ozo8bzpwPjwvbzpwPjwvcD4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6 bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20g Ni4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8ZGl2Pg0KPHAgY2xh c3M9Ik1zb05vcm1hbCI+SW4gb3ZpcnQgeW91IGhhdmUgdG8gYXR0YWNoIHN0b3JhZ2UgdGhyb3Vn aCBzcGVjaWZpYyBob3N0LiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAg Y2xhc3M9Ik1zb05vcm1hbCI+SWYgaG9zdCBnb2VzIGRvd24gc3RvcmFnZSBpcyBub3QgYXZhaWxh YmxlLiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29O b3JtYWwiPk9uIFNhdCwgQXByIDE1LCAyMDE3IGF0IDc6MzEgQU0gRkVSTkFORE8gRlJFRElBTkkg Jmx0OzxhIGhyZWY9Im1haWx0bzpmZXJuYW5kby5mcmVkaWFuaUB1cHguY29tIiB0YXJnZXQ9Il9i bGFuayI+ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbTwvYT4mZ3Q7IHdyb3RlOjxvOnA+PC9vOnA+ PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6 c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0 OjQuOHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCI+V2VsbCwgbWFrZSBpdCBub3QgZ28gdGhyb3VnaCBob3N0MSBhbmQgZGVkaWNh dGUgYSBzdG9yYWdlIHNlcnZlciBmb3IgcnVubmluZyBORlMgYW5kIG1ha2UgYm90aCBob3N0cyBj b25uZWN0IHRvIGl0LjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs IiBzdHlsZT0ibWFyZ2luLWJvdHRvbToxMi4wcHQiPkluIG15IHZpZXcgTkZTIGlzIG11Y2ggZWFz aWVyIHRvIG1hbmFnZSB0aGFuIGFueSBvdGhlciB0eXBlIG9mIHN0b3JhZ2UsIHNwZWNpYWxseSBG QyBhbmQgaVNDU0kgYW5kIHBlcmZvcm1hbmNlIGlzIHByZXR0eSBtdWNoIHRoZSBzYW1lLCBzbyB5 b3Ugd29uYHQgZ2V0IGJldHRlciByZXN1bHRzIG90aGVyIHRoYW4gbWFuYWdlbWVudCBnb2luZyB0 byBvdGhlciB0eXBlLjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs Ij5GZXJuYW5kbzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xh c3M9Ik1zb05vcm1hbCI+MjAxNy0wNC0xNSA1OjI1IEdNVC0wMzowMCBLb25zdGFudGluIFJhc2tv c2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzprb25yYXNrb0BnbWFpbC5jb20iIHRhcmdldD0iX2Js YW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+Jmd0Ozo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0K PC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3Jk ZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFy Z2luLWxlZnQ6NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+SGkgZ3V5cywgPG86cD48L286cD48L3A+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1h bCI+SSBoYXZlIG9uZSBuZnMmbmJzcDtzdG9yYWdlLDxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8 ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+aXQncyBjb25uZWN0ZWQgdGhyb3VnaCBob3N0MS48 bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPmhvc3Qy IGFsc28gaGFzIGFjY2VzcyB0byBpdCwgSSBjYW4gZWFzaWx5IG1pZ3JhdGUgdm1zJm5ic3A7YmV0 d2VlbiB0aGVtLjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNv Tm9ybWFsIj5UaGUgcXVlc3Rpb24gaXMgLSBpZiBob3N0MSBpcyBkb3duIC0gYWxsIGluZnJhc3Ry dWN0dXJlIGlzIGRvd24sIHNpbmNlIGFsbCB0cmFmZmljIGdvZXMgdGhyb3VnaCBob3N0MSw8bzpw PjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPmlzIHRoZXJl IGFueSB3YXkgaW4gb1ZpcnQgdG8gdXNlIHJlZHVuZGFudCBzdG9yYWdlPzxvOnA+PC9vOnA+PC9w Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48 L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5Pbmx5IGdsdXN0ZXJmcz88 bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+ Jm5ic3A7PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+VGhh bmtzPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48 bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs Ij48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8 ZGl2Pg0KPGRpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpz b2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6 NC44cHQ7bWFyZ2luLXJpZ2h0OjBjbSI+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibWFy Z2luLWJvdHRvbToxMi4wcHQiPl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fPGJyPg0KVXNlcnMgbWFpbGluZyBsaXN0PGJyPg0KPGEgaHJlZj0ibWFpbHRvOlVz ZXJzQG92aXJ0Lm9yZyIgdGFyZ2V0PSJfYmxhbmsiPlVzZXJzQG92aXJ0Lm9yZzwvYT48YnI+DQo8 YSBocmVmPSJodHRwOi8vbGlzdHMub3ZpcnQub3JnL21haWxtYW4vbGlzdGluZm8vdXNlcnMiIHRh cmdldD0iX2JsYW5rIj5odHRwOi8vbGlzdHMub3ZpcnQub3JnL21haWxtYW4vbGlzdGluZm8vdXNl cnM8L2E+PG86cD48L286cD48L3A+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9i bG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPHAgY2xh c3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjwvYmxvY2txdW90 ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIg c3R5bGU9Im1hcmdpbi1ib3R0b206MTIuMHB0Ij48YnI+DQpfX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fXzxicj4NClVzZXJzIG1haWxpbmcgbGlzdDxicj4NCjxh IGhyZWY9Im1haWx0bzpVc2Vyc0BvdmlydC5vcmciIHRhcmdldD0iX2JsYW5rIj5Vc2Vyc0Bvdmly dC5vcmc8L2E+PGJyPg0KPGEgaHJlZj0iaHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xp c3RpbmZvL3VzZXJzIiB0YXJnZXQ9Il9ibGFuayI+aHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWls bWFuL2xpc3RpbmZvL3VzZXJzPC9hPjxvOnA+PC9vOnA+PC9wPg0KPC9ibG9ja3F1b3RlPg0KPC9k aXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0K PHAgY2xhc3M9Ik1zb05vcm1hbCI+X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX188YnI+DQpVc2VycyBtYWlsaW5nIGxpc3Q8YnI+DQo8YSBocmVmPSJtYWlsdG86 VXNlcnNAb3ZpcnQub3JnIiB0YXJnZXQ9Il9ibGFuayI+VXNlcnNAb3ZpcnQub3JnPC9hPjxicj4N CjxhIGhyZWY9Imh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0aW5mby91c2VycyIg dGFyZ2V0PSJfYmxhbmsiPmh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0aW5mby91 c2VyczwvYT48bzpwPjwvbzpwPjwvcD4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8 L2Rpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtYXJnaW4tYm90dG9tOjEyLjBwdCI+ PGJyPg0KX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX188YnI+ DQpVc2VycyBtYWlsaW5nIGxpc3Q8YnI+DQo8YSBocmVmPSJtYWlsdG86VXNlcnNAb3ZpcnQub3Jn IiB0YXJnZXQ9Il9ibGFuayI+VXNlcnNAb3ZpcnQub3JnPC9hPjxicj4NCjxhIGhyZWY9Imh0dHA6 Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0aW5mby91c2VycyIgdGFyZ2V0PSJfYmxhbmsi Pmh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0aW5mby91c2VyczwvYT48bzpwPjwv bzpwPjwvcD4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86 cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+ DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0K PC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwv ZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ibG9ja3F1b3Rl Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4N CjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8 L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90 ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+ DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9i bG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ib2R5Pg0KPC9odG1sPg0K --_000_EF8A6CD043B349ADB14F59EB42E5B511acroniscom_--
On Mon, Apr 17, 2017 at 9:26 AM, Pavel Gashev <Pax@acronis.com> wrote:
Nir,
Isn’t SPM managed via Sanlock? I believe there is no need to fence SPM host. Especially if there are no SPM tasks running.
It's true that the exclusivity of the SPM role is enforced by Sanlock but you always need to fence a non-responsive SPM because there is no way to guarantee that the host is not still manipulating storage (eg. LV extensions) and we must ensure that only one host has the masterfs on the master storage domain mounted.
*From: *<users-bounces@ovirt.org> on behalf of Nir Soffer < nsoffer@redhat.com> *Date: *Monday, 17 April 2017 at 16:06 *To: *Konstantin Raskoshnyi <konrasko@gmail.com>, Dan Yasny < dyasny@gmail.com> *Cc: *users <users@ovirt.org>, FERNANDO FREDIANI < fernando.frediani@upx.com> *Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1'
2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5'
2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you
configure power management for all hosts? did you check that it
work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot
move the spm to another host, unless you manually confirm that the
SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4,
Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense.
I was trying to set it up, but doesn't work with our staging hardware.
We have old ilo100, I'll try again.
Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"?
I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host
will not move to a new host until the spm host is online again, or you confirm
manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start
the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Hi Fernando,
I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.
Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host.
If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.
In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys,
I have one nfs storage,
it's connected through host1.
host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,
is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
-- Adam Litke
--_000_87040B69038146B985204DA5C6360EEAacroniscom_ Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 QWRhbSwNCg0KWW91IGtub3csIFNhbmxvY2sgaGFzIHJlY292ZXJ5IG1lY2hhbmlzbSB0aGF0IGtp bGxzIFZEU00sIG9yIGV2ZW4gdHJpZ2dlcnMgV2F0Y2hkb2cgdG8gcmVib290IFNQTSBob3N0IGlu IGNhc2UgaXQgaGFzIGxvc3QgdGhlIFNQTSBsb2NrLg0KSeKAmW0gYXNraW5nIGJlY2F1c2UgSSBo YWQgaXNzdWVzIHdpdGggbXkgbWFzdGVyIHN0b3JhZ2UgdGhhdCBjYXVzZWQgU1BNIGhvc3QgdG8g cmVib290IGJ5IFdhdGNoZG9nLiBBbmQgSSB3YXMgc3VyZSB0aGF0IGl04oCZcyBhbiBpbnRlbmRl ZCBiZWhhdmlvdXIuIElzbuKAmXQgaXQ/DQoNCg0KRnJvbTogQWRhbSBMaXRrZSA8YWxpdGtlQHJl ZGhhdC5jb20+DQpEYXRlOiBNb25kYXksIDE3IEFwcmlsIDIwMTcgYXQgMTc6MzINClRvOiBQYXZl bCBHYXNoZXYgPFBheEBhY3JvbmlzLmNvbT4NCkNjOiBOaXIgU29mZmVyIDxuc29mZmVyQHJlZGhh dC5jb20+LCB1c2VycyA8dXNlcnNAb3ZpcnQub3JnPg0KU3ViamVjdDogUmU6IFtvdmlydC11c2Vy c10gc3RvcmFnZSByZWR1bmRhbmN5IGluIE92aXJ0DQoNCg0KDQpPbiBNb24sIEFwciAxNywgMjAx NyBhdCA5OjI2IEFNLCBQYXZlbCBHYXNoZXYgPFBheEBhY3JvbmlzLmNvbTxtYWlsdG86UGF4QGFj cm9uaXMuY29tPj4gd3JvdGU6DQpOaXIsDQoNCklzbuKAmXQgU1BNIG1hbmFnZWQgdmlhIFNhbmxv Y2s/IEkgYmVsaWV2ZSB0aGVyZSBpcyBubyBuZWVkIHRvIGZlbmNlIFNQTSBob3N0LiBFc3BlY2lh bGx5IGlmIHRoZXJlIGFyZSBubyBTUE0gdGFza3MgcnVubmluZy4NCg0KSXQncyB0cnVlIHRoYXQg dGhlIGV4Y2x1c2l2aXR5IG9mIHRoZSBTUE0gcm9sZSBpcyBlbmZvcmNlZCBieSBTYW5sb2NrIGJ1 dCB5b3UgYWx3YXlzIG5lZWQgdG8gZmVuY2UgYSBub24tcmVzcG9uc2l2ZSBTUE0gYmVjYXVzZSB0 aGVyZSBpcyBubyB3YXkgdG8gZ3VhcmFudGVlIHRoYXQgdGhlIGhvc3QgaXMgbm90IHN0aWxsIG1h bmlwdWxhdGluZyBzdG9yYWdlIChlZy4gTFYgZXh0ZW5zaW9ucykgYW5kIHdlIG11c3QgZW5zdXJl IHRoYXQgb25seSBvbmUgaG9zdCBoYXMgdGhlIG1hc3RlcmZzIG9uIHRoZSBtYXN0ZXIgc3RvcmFn ZSBkb21haW4gbW91bnRlZC4NCg0KDQoNCkZyb206IDx1c2Vycy1ib3VuY2VzQG92aXJ0Lm9yZzxt YWlsdG86dXNlcnMtYm91bmNlc0BvdmlydC5vcmc+PiBvbiBiZWhhbGYgb2YgTmlyIFNvZmZlciA8 bnNvZmZlckByZWRoYXQuY29tPG1haWx0bzpuc29mZmVyQHJlZGhhdC5jb20+Pg0KRGF0ZTogTW9u ZGF5LCAxNyBBcHJpbCAyMDE3IGF0IDE2OjA2DQpUbzogS29uc3RhbnRpbiBSYXNrb3NobnlpIDxr b25yYXNrb0BnbWFpbC5jb208bWFpbHRvOmtvbnJhc2tvQGdtYWlsLmNvbT4+LCBEYW4gWWFzbnkg PGR5YXNueUBnbWFpbC5jb208bWFpbHRvOmR5YXNueUBnbWFpbC5jb20+Pg0KQ2M6IHVzZXJzIDx1 c2Vyc0BvdmlydC5vcmc8bWFpbHRvOnVzZXJzQG92aXJ0Lm9yZz4+LCBGRVJOQU5ETyBGUkVESUFO SSA8ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbTxtYWlsdG86ZmVybmFuZG8uZnJlZGlhbmlAdXB4 LmNvbT4+DQpTdWJqZWN0OiBSZTogW292aXJ0LXVzZXJzXSBzdG9yYWdlIHJlZHVuZGFuY3kgaW4g T3ZpcnQNCg0KT24gTW9uLCBBcHIgMTcsIDIwMTcgYXQgODoyNCBBTSBLb25zdGFudGluIFJhc2tv c2hueWkgPGtvbnJhc2tvQGdtYWlsLmNvbTxtYWlsdG86a29ucmFza29AZ21haWwuY29tPj4gd3Jv dGU6DQpCdXQgYWN0dWFsbHksIGl0IGRpZG4ndCB3b3JrIHdlbGwuIEFmdGVyIG1haW4gU1BNIGhv c3Qgd2VudCBkb3duIEkgc2VlIHRoaXMNCg0KDQoyMDE3LTA0LTE3IDA1OjIzOjE1LDU1NFogRVJS T1IgW29yZy5vdmlydC5lbmdpbmUuY29yZS52ZHNicm9rZXIuaXJzYnJva2VyLklyc1Byb3h5XSAo RGVmYXVsdFF1YXJ0elNjaGVkdWxlcjUpIFs0ZGNjMDMzZC0yNmJmLTQ5YmItYmZhYS0wM2E5NzBk YmJlYzFdIFNQTSBJbml0OiBjb3VsZCBub3QgZmluZCByZXBvcnRlZCB2ZHMgb3Igbm90IHVwIC0g cG9vbDogJ1NURycgdmRzX3NwbV9pZDogJzEnDQoyMDE3LTA0LTE3IDA1OjIzOjE1LDU2N1ogSU5G TyAgW29yZy5vdmlydC5lbmdpbmUuY29yZS52ZHNicm9rZXIuaXJzYnJva2VyLklyc1Byb3h5XSAo RGVmYXVsdFF1YXJ0elNjaGVkdWxlcjUpIFs0ZGNjMDMzZC0yNmJmLTQ5YmItYmZhYS0wM2E5NzBk YmJlYzFdIFNQTSBzZWxlY3Rpb24gLSB2ZHMgc2VlbXMgYXMgc3BtICd0YW5rNScNCjIwMTctMDQt MTcgMDU6MjM6MTUsNTY3WiBXQVJOICBbb3JnLm92aXJ0LmVuZ2luZS5jb3JlLnZkc2Jyb2tlci5p cnNicm9rZXIuSXJzUHJveHldIChEZWZhdWx0UXVhcnR6U2NoZWR1bGVyNSkgWzRkY2MwMzNkLTI2 YmYtNDliYi1iZmFhLTAzYTk3MGRiYmVjMV0gc3BtIHZkcyBpcyBub24gcmVzcG9uc2l2ZSwgc3Rv cHBpbmcgc3BtIHNlbGVjdGlvbi4NCg0KU28gdGhhdCBtZWFucyBvbmx5IGlmIEJNQyBpcyB1cCBp dCdzIHBvc3NpYmxlIHRvIGF1dG9tYXRpY2FsbHkgc3dpdGNoICBTUE0gaG9zdD8NCg0KQk1DPw0K DQpJZiB5b3VyIFNQTSBpcyBubyByZXNwb25zaXZlLCB0aGUgc3lzdGVtIHdpbGwgdHJ5IHRvIGZl bmNlIGl0LiBEaWQgeW91DQpjb25maWd1cmUgcG93ZXIgbWFuYWdlbWVudCBmb3IgYWxsIGhvc3Rz PyBkaWQgeW91IGNoZWNrIHRoYXQgaXQNCndvcms/IEhvdyBkaWQgeW91IHNpbXVsYXRlIG5vbi1y ZXNwb25zaXZlIGhvc3Q/DQoNCklmIHBvd2VyIG1hbmFnZW1lbnQgaXMgbm90IGNvbmZpZ3VyZWQg b3IgZmFpbCwgdGhlIHN5c3RlbSBjYW5ub3QNCm1vdmUgdGhlIHNwbSB0byBhbm90aGVyIGhvc3Qs IHVubGVzcyB5b3UgbWFudWFsbHkgY29uZmlybSB0aGF0IHRoZQ0KU1BNIGhvc3Qgd2FzIHJlYm9v dGVkLg0KDQpOaXINCg0KDQpUaGFua3MNCg0KT24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgODoyOSBQ TSwgS29uc3RhbnRpbiBSYXNrb3NobnlpIDxrb25yYXNrb0BnbWFpbC5jb208bWFpbHRvOmtvbnJh c2tvQGdtYWlsLmNvbT4+IHdyb3RlOg0KT2gsIGZlbmNlIGFnZW50IHdvcmtzIGZpbmUgaWYgSSBz ZWxlY3QgaWxvNCwNClRoYW5rIHlvdSBmb3IgeW91ciBoZWxwIQ0KDQpPbiBTdW4sIEFwciAxNiwg MjAxNyBhdCA4OjIyIFBNIERhbiBZYXNueSA8ZHlhc255QGdtYWlsLmNvbTxtYWlsdG86ZHlhc255 QGdtYWlsLmNvbT4+IHdyb3RlOg0KT24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgMTE6MTkgUE0sIEtv bnN0YW50aW4gUmFza29zaG55aSA8a29ucmFza29AZ21haWwuY29tPG1haWx0bzprb25yYXNrb0Bn bWFpbC5jb20+PiB3cm90ZToNCk1ha2VzIHNlbnNlLg0KSSB3YXMgdHJ5aW5nIHRvIHNldCBpdCB1 cCwgYnV0IGRvZXNuJ3Qgd29yayB3aXRoIG91ciBzdGFnaW5nIGhhcmR3YXJlLg0KV2UgaGF2ZSBv bGQgaWxvMTAwLCBJJ2xsIHRyeSBhZ2Fpbi4NClRoYW5rcyENCg0KDQpJdCBpcyBhYnNvbHV0ZWx5 IG5lY2Vzc2FyeSBmb3IgYW55IEhBIHRvIHdvcmsgcHJvcGVybHkuIFRoZXJlJ3Mgb2YgY291cnNl IHRoZSAiY29uZmlybSBob3N0IGhhcyBiZWVuIHNodXRkb3duIiBvcHRpb24sIHdoaWNoIHNlcnZl cyBhcyBhbiBvdmVycmlkZSBmb3IgdGhlIGZlbmNlIGNvbW1hbmQsIGJ1dCBpdCdzIG1hbnVhbA0K DQpPbiBTdW4sIEFwciAxNiwgMjAxNyBhdCA4OjE4IFBNIERhbiBZYXNueSA8ZHlhc255QGdtYWls LmNvbTxtYWlsdG86ZHlhc255QGdtYWlsLmNvbT4+IHdyb3RlOg0KT24gU3VuLCBBcHIgMTYsIDIw MTcgYXQgMTE6MTUgUE0sIEtvbnN0YW50aW4gUmFza29zaG55aSA8a29ucmFza29AZ21haWwuY29t PG1haWx0bzprb25yYXNrb0BnbWFpbC5jb20+PiB3cm90ZToNCkZlbmNlIGFnZW50IHVuZGVyIGVh Y2ggbm9kZT8NCg0KV2hlbiB5b3UgY29uZmlndXJlIGEgaG9zdCwgdGhlcmUncyB0aGUgcG93ZXIg bWFuYWdlbWVudCB0YWIsIHdoZXJlIHlvdSBuZWVkIHRvIGVudGVyIHRoZSBibWMgZGV0YWlscyBm b3IgdGhlIGhvc3QuIElmIHlvdSBkb24ndCBoYXZlIGZlbmNpbmcgZW5hYmxlZCwgaG93IGRvIHlv dSBleHBlY3QgdGhlIHN5c3RlbSB0byBtYWtlIHN1cmUgYSBob3N0IHJ1bm5pbmcgYSBzZXJ2aWNl IGlzIGFjdHVhbGx5IGRvd24gKGFuZCBpdCBpcyBzYWZlIHRvIHN0YXJ0IEhBIHNlcnZpY2VzIGVs c2V3aGVyZSksIGFuZCBub3QsIGZvciBleGFtcGxlLCBqdXN0IHVucmVhY2hhYmxlIGJ5IHRoZSBl bmdpbmU/IEhvdyBkbyB5b3UgYXZvaWQgYSBzcGxpdGJyYWlkIC0+IFNCQSA/DQoNCg0KT24gU3Vu LCBBcHIgMTYsIDIwMTcgYXQgODoxNCBQTSBEYW4gWWFzbnkgPGR5YXNueUBnbWFpbC5jb208bWFp bHRvOmR5YXNueUBnbWFpbC5jb20+PiB3cm90ZToNCk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDEx OjEzIFBNLCBLb25zdGFudGluIFJhc2tvc2hueWkgPGtvbnJhc2tvQGdtYWlsLmNvbTxtYWlsdG86 a29ucmFza29AZ21haWwuY29tPj4gd3JvdGU6DQoiQ29ybmVyIGNhc2VzIj8NCkkgdHJpZWQgdG8g c2ltdWxhdGUgY3Jhc2ggb2YgU1BNIHNlcnZlciBhbmQgb3ZpcnQga2VwdCB0cnlpbmcgdG8gcmVp c3RhYmxpc2hlZCBjb25uZWN0aW9uIHRvIHRoZSBmYWlsZWQgbm9kZS4NCg0KRGlkIHlvdSBjb25m aWd1cmUgZmVuY2luZz8NCg0KDQoNCk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDg6MTAgUE0gRGFu IFlhc255IDxkeWFzbnlAZ21haWwuY29tPG1haWx0bzpkeWFzbnlAZ21haWwuY29tPj4gd3JvdGU6 DQpPbiBTdW4sIEFwciAxNiwgMjAxNyBhdCA3OjI5IEFNLCBOaXIgU29mZmVyIDxuc29mZmVyQHJl ZGhhdC5jb208bWFpbHRvOm5zb2ZmZXJAcmVkaGF0LmNvbT4+IHdyb3RlOg0KT24gU3VuLCBBcHIg MTYsIDIwMTcgYXQgMjowNSBQTSBEYW4gWWFzbnkgPGR5YXNueUByZWRoYXQuY29tPG1haWx0bzpk eWFzbnlAcmVkaGF0LmNvbT4+IHdyb3RlOg0KDQoNCk9uIEFwciAxNiwgMjAxNyA3OjAxIEFNLCAi TmlyIFNvZmZlciIgPG5zb2ZmZXJAcmVkaGF0LmNvbTxtYWlsdG86bnNvZmZlckByZWRoYXQuY29t Pj4gd3JvdGU6DQpPbiBTdW4sIEFwciAxNiwgMjAxNyBhdCA0OjE3IEFNIERhbiBZYXNueSA8ZHlh c255QGdtYWlsLmNvbTxtYWlsdG86ZHlhc255QGdtYWlsLmNvbT4+IHdyb3RlOg0KV2hlbiB5b3Ug c2V0IHVwIGEgc3RvcmFnZSBkb21haW4sIHlvdSBuZWVkIHRvIHNwZWNpZnkgYSBob3N0IHRvIHBl cmZvcm0gdGhlIGluaXRpYWwgc3RvcmFnZSBvcGVyYXRpb25zLCBidXQgb25jZSB0aGUgU0QgaXMg ZGVmaW5lZCwgaXQncyBkZXRhaWxzIGFyZSBpbiB0aGUgZW5naW5lIGRhdGFiYXNlLCBhbmQgYWxs IHRoZSBob3N0cyBnZXQgY29ubmVjdGVkIHRvIGl0IGRpcmVjdGx5LiBJZiB0aGUgZmlyc3QgaG9z dCB5b3UgdXNlZCB0byBkZWZpbmUgdGhlIFNEIGdvZXMgZG93biwgYWxsIG90aGVyIGhvc3RzIHdp bGwgc3RpbGwgcmVtYWluIGNvbm5lY3RlZCBhbmQgd29yay4gU1BNIGlzIGFuIEhBIHNlcnZpY2Us IGFuZCBpZiB0aGUgY3VycmVudCBTUE0gaG9zdCBnb2VzIGRvd24sIFNQTSBnZXRzIHN0YXJ0ZWQg b24gYW5vdGhlciBob3N0IGluIHRoZSBEQy4gSW4gc2hvcnQsIHVubGVzcyB5b3VyIGFjdHVhbCBO RlMgZXhwb3J0aW5nIGhvc3QgZ29lcyBkb3duLCB0aGVyZSBpcyBubyBvdXRhZ2UuDQoNClRoZXJl IGlzIG5vIHN0b3JhZ2Ugb3V0YWdlLCBidXQgaWYgeW91IHNodXRkb3duIHRoZSBzcG0gaG9zdCwg dGhlIHNwbSBob3N0DQp3aWxsIG5vdCBtb3ZlIHRvIGEgbmV3IGhvc3QgdW50aWwgdGhlIHNwbSBo b3N0IGlzIG9ubGluZSBhZ2Fpbiwgb3IgeW91IGNvbmZpcm0NCm1hbnVhbGx5IHRoYXQgdGhlIHNw bSBob3N0IHdhcyByZWJvb3RlZC4NCg0KSW4gYSBwcm9wZXJseSBjb25maWd1cmVkIHNldHVwIHRo ZSBTQkEgc2hvdWxkIHRha2UgY2FyZSBvZiB0aGF0LiBUaGF0J3MgdGhlIHdob2xlIHBvaW50IG9m IEhBIHNlcnZpY2VzDQoNCkluIHNvbWUgY2FzZXMgbGlrZSBwb3dlciBsb3NzIG9yIGhhcmR3YXJl IGZhaWx1cmUsIHRoZXJlIGlzIG5vIHdheSB0byBzdGFydA0KdGhlIHNwbSBob3N0LCBhbmQgdGhl IHN5c3RlbSBjYW5ub3QgcmVjb3ZlciBhdXRvbWF0aWNhbGx5Lg0KDQpUaGVyZSBhcmUgYWx3YXlz IGNvcm5lciBjYXNlcywgbm8gZG91YnQuIEJ1dCBpbiBhIG5vcm1hbCBzaXR1YXRpb24uIHdoZXJl IGFuIFNQTSBob3N0IGdvZXMgZG93biBiZWNhdXNlIG9mIGEgaGFyZHdhcmUgZmFpbHVyZSwgaXQg Z2V0cyBmZW5jZWQsIG90aGVyIGhvc3RzIGNvbnRlbmQgZm9yIFNQTSBhbmQgc3RhcnQgaXQuIE5v IHN1cnByaXNlcyB0aGVyZS4NCg0KDQpOaXINCg0KDQoNCk5pcg0KDQoNCk9uIFNhdCwgQXByIDE1 LCAyMDE3IGF0IDE6NTMgUE0sIEtvbnN0YW50aW4gUmFza29zaG55aSA8a29ucmFza29AZ21haWwu Y29tPG1haWx0bzprb25yYXNrb0BnbWFpbC5jb20+PiB3cm90ZToNCkhpIEZlcm5hbmRvLA0KSSBz ZWUgZWFjaCBob3N0IGhhcyBkaXJlY3QgY29ubmVjdGlvbiBuZnMgbW91bnQsIGJ1dCB5ZXMsIGlm IG1haW4gaG9zdCB0byB3aGljaCBJIGNvbm5lY3RlZCBuZnMgc3RvcmFnZSBnb2luZyBkb3duIHRo ZSBzdG9yYWdlIGJlY29tZXMgdW5hdmFpbGFibGUgYW5kIGFsbCB2bXMgYXJlIGRvd24NCg0KDQpP biBTYXQsIEFwciAxNSwgMjAxNyBhdCAxMDozNyBBTSBGRVJOQU5ETyBGUkVESUFOSSA8ZmVybmFu ZG8uZnJlZGlhbmlAdXB4LmNvbTxtYWlsdG86ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbT4+IHdy b3RlOg0KSGVsbG8gS29uc3RhbnRpbi4NClRoYXQgZG9lc25gdCBtYWtlIG11Y2ggc2Vuc2UgbWFr ZSBhIHdob2xlIGNsdXN0ZXIgZGVwZW5kIG9uIGEgc2luZ2xlIGhvc3QuIEZyb20gd2hhdCBJIGtu b3cgYW55IGhvc3QgdGFsayBkaXJlY3RseSB0byBORlMgU3RvcmFnZSBBcnJheSBvciB3aGF0ZXZl ciBvdGhlciBTaGFyZWQgU3RvcmFnZSB5b3UgaGF2ZS4NCkhhdmUgeW91IHRlc3RlZCB0aGF0IGhv c3QgZ29pbmcgZG93biBpZiB0aGF0IGFmZmVjdHMgdGhlIG90aGVyIHdpdGggdGhlIE5GUyBtb3Vu dGVkIGRpcmVjdGxseSBpbiBhIE5GUyBTdG9yYWdlIGFycmF5ID8NCkZlcm5hbmRvDQoNCjIwMTct MDQtMTUgMTI6NDIgR01ULTAzOjAwIEtvbnN0YW50aW4gUmFza29zaG55aSA8a29ucmFza29AZ21h aWwuY29tPG1haWx0bzprb25yYXNrb0BnbWFpbC5jb20+PjoNCkluIG92aXJ0IHlvdSBoYXZlIHRv IGF0dGFjaCBzdG9yYWdlIHRocm91Z2ggc3BlY2lmaWMgaG9zdC4NCklmIGhvc3QgZ29lcyBkb3du IHN0b3JhZ2UgaXMgbm90IGF2YWlsYWJsZS4NCg0KT24gU2F0LCBBcHIgMTUsIDIwMTcgYXQgNzoz MSBBTSBGRVJOQU5ETyBGUkVESUFOSSA8ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbTxtYWlsdG86 ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbT4+IHdyb3RlOg0KV2VsbCwgbWFrZSBpdCBub3QgZ28g dGhyb3VnaCBob3N0MSBhbmQgZGVkaWNhdGUgYSBzdG9yYWdlIHNlcnZlciBmb3IgcnVubmluZyBO RlMgYW5kIG1ha2UgYm90aCBob3N0cyBjb25uZWN0IHRvIGl0Lg0KSW4gbXkgdmlldyBORlMgaXMg bXVjaCBlYXNpZXIgdG8gbWFuYWdlIHRoYW4gYW55IG90aGVyIHR5cGUgb2Ygc3RvcmFnZSwgc3Bl Y2lhbGx5IEZDIGFuZCBpU0NTSSBhbmQgcGVyZm9ybWFuY2UgaXMgcHJldHR5IG11Y2ggdGhlIHNh bWUsIHNvIHlvdSB3b25gdCBnZXQgYmV0dGVyIHJlc3VsdHMgb3RoZXIgdGhhbiBtYW5hZ2VtZW50 IGdvaW5nIHRvIG90aGVyIHR5cGUuDQpGZXJuYW5kbw0KDQoyMDE3LTA0LTE1IDU6MjUgR01ULTAz OjAwIEtvbnN0YW50aW4gUmFza29zaG55aSA8a29ucmFza29AZ21haWwuY29tPG1haWx0bzprb25y YXNrb0BnbWFpbC5jb20+PjoNCkhpIGd1eXMsDQpJIGhhdmUgb25lIG5mcyBzdG9yYWdlLA0KaXQn cyBjb25uZWN0ZWQgdGhyb3VnaCBob3N0MS4NCmhvc3QyIGFsc28gaGFzIGFjY2VzcyB0byBpdCwg SSBjYW4gZWFzaWx5IG1pZ3JhdGUgdm1zIGJldHdlZW4gdGhlbS4NCg0KVGhlIHF1ZXN0aW9uIGlz IC0gaWYgaG9zdDEgaXMgZG93biAtIGFsbCBpbmZyYXN0cnVjdHVyZSBpcyBkb3duLCBzaW5jZSBh bGwgdHJhZmZpYyBnb2VzIHRocm91Z2ggaG9zdDEsDQppcyB0aGVyZSBhbnkgd2F5IGluIG9WaXJ0 IHRvIHVzZSByZWR1bmRhbnQgc3RvcmFnZT8NCg0KT25seSBnbHVzdGVyZnM/DQoNClRoYW5rcw0K DQoNCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQpVc2Vy cyBtYWlsaW5nIGxpc3QNClVzZXJzQG92aXJ0Lm9yZzxtYWlsdG86VXNlcnNAb3ZpcnQub3JnPg0K aHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL3VzZXJzDQoNCg0KX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NClVzZXJzIG1haWxpbmcg bGlzdA0KVXNlcnNAb3ZpcnQub3JnPG1haWx0bzpVc2Vyc0BvdmlydC5vcmc+DQpodHRwOi8vbGlz dHMub3ZpcnQub3JnL21haWxtYW4vbGlzdGluZm8vdXNlcnMNCg0KX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX18NClVzZXJzIG1haWxpbmcgbGlzdA0KVXNlcnNA b3ZpcnQub3JnPG1haWx0bzpVc2Vyc0BvdmlydC5vcmc+DQpodHRwOi8vbGlzdHMub3ZpcnQub3Jn L21haWxtYW4vbGlzdGluZm8vdXNlcnMNCg0KX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX18NClVzZXJzIG1haWxpbmcgbGlzdA0KVXNlcnNAb3ZpcnQub3JnPG1h aWx0bzpVc2Vyc0BvdmlydC5vcmc+DQpodHRwOi8vbGlzdHMub3ZpcnQub3JnL21haWxtYW4vbGlz dGluZm8vdXNlcnMNCg0KDQoNCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fDQpVc2VycyBtYWlsaW5nIGxpc3QNClVzZXJzQG92aXJ0Lm9yZzxtYWlsdG86VXNl cnNAb3ZpcnQub3JnPg0KaHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL3Vz ZXJzDQoNCg0KDQotLQ0KQWRhbSBMaXRrZQ0K --_000_87040B69038146B985204DA5C6360EEAacroniscom_ Content-Type: text/html; charset="utf-8" Content-ID: <B553B5567B282947AF72F9C192AD4EE5@acronis.com> Content-Transfer-Encoding: base64 PGh0bWwgeG1sbnM6bz0idXJuOnNjaGVtYXMtbWljcm9zb2Z0LWNvbTpvZmZpY2U6b2ZmaWNlIiB4 bWxuczp3PSJ1cm46c2NoZW1hcy1taWNyb3NvZnQtY29tOm9mZmljZTp3b3JkIiB4bWxuczptPSJo dHRwOi8vc2NoZW1hcy5taWNyb3NvZnQuY29tL29mZmljZS8yMDA0LzEyL29tbWwiIHhtbG5zPSJo dHRwOi8vd3d3LnczLm9yZy9UUi9SRUMtaHRtbDQwIj4NCjxoZWFkPg0KPG1ldGEgaHR0cC1lcXVp dj0iQ29udGVudC1UeXBlIiBjb250ZW50PSJ0ZXh0L2h0bWw7IGNoYXJzZXQ9dXRmLTgiPg0KPG1l dGEgbmFtZT0iVGl0bGUiIGNvbnRlbnQ9IiI+DQo8bWV0YSBuYW1lPSJLZXl3b3JkcyIgY29udGVu dD0iIj4NCjxtZXRhIG5hbWU9IkdlbmVyYXRvciIgY29udGVudD0iTWljcm9zb2Z0IFdvcmQgMTUg KGZpbHRlcmVkIG1lZGl1bSkiPg0KPHN0eWxlPjwhLS0NCi8qIEZvbnQgRGVmaW5pdGlvbnMgKi8N CkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6IkNhbWJyaWEgTWF0aCI7DQoJcGFub3NlLTE6MiA0 IDUgMyA1IDQgNiAzIDIgNDt9DQpAZm9udC1mYWNlDQoJe2ZvbnQtZmFtaWx5OkNhbGlicmk7DQoJ cGFub3NlLTE6MiAxNSA1IDIgMiAyIDQgMyAyIDQ7fQ0KLyogU3R5bGUgRGVmaW5pdGlvbnMgKi8N CnAuTXNvTm9ybWFsLCBsaS5Nc29Ob3JtYWwsIGRpdi5Nc29Ob3JtYWwNCgl7bWFyZ2luOjBjbTsN CgltYXJnaW4tYm90dG9tOi4wMDAxcHQ7DQoJZm9udC1zaXplOjEyLjBwdDsNCglmb250LWZhbWls eToiVGltZXMgTmV3IFJvbWFuIjt9DQphOmxpbmssIHNwYW4uTXNvSHlwZXJsaW5rDQoJe21zby1z dHlsZS1wcmlvcml0eTo5OTsNCgljb2xvcjpibHVlOw0KCXRleHQtZGVjb3JhdGlvbjp1bmRlcmxp bmU7fQ0KYTp2aXNpdGVkLCBzcGFuLk1zb0h5cGVybGlua0ZvbGxvd2VkDQoJe21zby1zdHlsZS1w cmlvcml0eTo5OTsNCgljb2xvcjpwdXJwbGU7DQoJdGV4dC1kZWNvcmF0aW9uOnVuZGVybGluZTt9 DQpwDQoJe21zby1zdHlsZS1wcmlvcml0eTo5OTsNCgltc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzsN CgltYXJnaW4tcmlnaHQ6MGNtOw0KCW1zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvOw0KCW1hcmdp bi1sZWZ0OjBjbTsNCglmb250LXNpemU6MTIuMHB0Ow0KCWZvbnQtZmFtaWx5OiJUaW1lcyBOZXcg Um9tYW4iO30NCnNwYW4uRW1haWxTdHlsZTE4DQoJe21zby1zdHlsZS10eXBlOnBlcnNvbmFsLXJl cGx5Ow0KCWZvbnQtZmFtaWx5OkNhbGlicmk7DQoJY29sb3I6d2luZG93dGV4dDt9DQpzcGFuLm1z b0lucw0KCXttc28tc3R5bGUtdHlwZTpleHBvcnQtb25seTsNCgltc28tc3R5bGUtbmFtZToiIjsN Cgl0ZXh0LWRlY29yYXRpb246dW5kZXJsaW5lOw0KCWNvbG9yOnRlYWw7fQ0KLk1zb0NocERlZmF1 bHQNCgl7bXNvLXN0eWxlLXR5cGU6ZXhwb3J0LW9ubHk7DQoJZm9udC1zaXplOjEwLjBwdDt9DQpA cGFnZSBXb3JkU2VjdGlvbjENCgl7c2l6ZTo1OTUuMHB0IDg0Mi4wcHQ7DQoJbWFyZ2luOjIuMGNt IDQyLjVwdCAyLjBjbSAzLjBjbTt9DQpkaXYuV29yZFNlY3Rpb24xDQoJe3BhZ2U6V29yZFNlY3Rp b24xO30NCi0tPjwvc3R5bGU+DQo8L2hlYWQ+DQo8Ym9keSBiZ2NvbG9yPSJ3aGl0ZSIgbGFuZz0i RU4tR0IiIGxpbms9ImJsdWUiIHZsaW5rPSJwdXJwbGUiPg0KPGRpdiBjbGFzcz0iV29yZFNlY3Rp b24xIj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEuMHB0 O2ZvbnQtZmFtaWx5OkNhbGlicmk7bXNvLWZhcmVhc3QtbGFuZ3VhZ2U6RU4tVVMiPkFkYW0sPG86 cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZv bnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6Q2FsaWJyaTttc28tZmFyZWFzdC1sYW5ndWFnZTpF Ti1VUyI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+ PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6Q2FsaWJyaTttc28tZmFy ZWFzdC1sYW5ndWFnZTpFTi1VUyI+WW91IGtub3csIFNhbmxvY2sgaGFzIHJlY292ZXJ5IG1lY2hh bmlzbSB0aGF0IGtpbGxzIFZEU00sIG9yIGV2ZW4gdHJpZ2dlcnMgV2F0Y2hkb2cgdG8gcmVib290 IFNQTSBob3N0IGluIGNhc2UgaXQgaGFzIGxvc3QgdGhlIFNQTSBsb2NrLjxvOnA+PC9vOnA+PC9z cGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEu MHB0O2ZvbnQtZmFtaWx5OkNhbGlicmk7bXNvLWZhcmVhc3QtbGFuZ3VhZ2U6RU4tVVMiPknigJlt IGFza2luZyBiZWNhdXNlIEkgaGFkIGlzc3VlcyB3aXRoIG15IG1hc3RlciBzdG9yYWdlIHRoYXQg Y2F1c2VkIFNQTSBob3N0IHRvIHJlYm9vdCBieSBXYXRjaGRvZy4gQW5kIEkgd2FzIHN1cmUgdGhh dCBpdOKAmXMgYW4gaW50ZW5kZWQgYmVoYXZpb3VyLiBJc27igJl0IGl0PzxvOnA+PC9vOnA+PC9z cGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEu MHB0O2ZvbnQtZmFtaWx5OkNhbGlicmk7bXNvLWZhcmVhc3QtbGFuZ3VhZ2U6RU4tVVMiPjxvOnA+ Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxl PSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OkNhbGlicmk7bXNvLWZhcmVhc3QtbGFuZ3Vh Z2U6RU4tVVMiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxkaXYgc3R5bGU9ImJvcmRl cjpub25lO2JvcmRlci10b3A6c29saWQgI0I1QzRERiAxLjBwdDtwYWRkaW5nOjMuMHB0IDBjbSAw Y20gMGNtIj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxiPjxzcGFuIHN0eWxlPSJmb250LWZhbWls eTpDYWxpYnJpO2NvbG9yOmJsYWNrIj5Gcm9tOiA8L3NwYW4+DQo8L2I+PHNwYW4gc3R5bGU9ImZv bnQtZmFtaWx5OkNhbGlicmk7Y29sb3I6YmxhY2siPkFkYW0gTGl0a2UgJmx0O2FsaXRrZUByZWRo YXQuY29tJmd0Ozxicj4NCjxiPkRhdGU6IDwvYj5Nb25kYXksIDE3IEFwcmlsIDIwMTcgYXQgMTc6 MzI8YnI+DQo8Yj5UbzogPC9iPlBhdmVsIEdhc2hldiAmbHQ7UGF4QGFjcm9uaXMuY29tJmd0Ozxi cj4NCjxiPkNjOiA8L2I+TmlyIFNvZmZlciAmbHQ7bnNvZmZlckByZWRoYXQuY29tJmd0OywgdXNl cnMgJmx0O3VzZXJzQG92aXJ0Lm9yZyZndDs8YnI+DQo8Yj5TdWJqZWN0OiA8L2I+UmU6IFtvdmly dC11c2Vyc10gc3RvcmFnZSByZWR1bmRhbmN5IGluIE92aXJ0PG86cD48L286cD48L3NwYW4+PC9w Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8L286cD48 L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpw PjwvcD4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4N CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5PbiBNb24sIEFwciAxNywgMjAxNyBhdCA5OjI2 IEFNLCBQYXZlbCBHYXNoZXYgJmx0OzxhIGhyZWY9Im1haWx0bzpQYXhAYWNyb25pcy5jb20iIHRh cmdldD0iX2JsYW5rIj5QYXhAYWNyb25pcy5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwv cD4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0ND Q0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFy Z2luLXJpZ2h0OjBjbSI+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNw YW4gc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6Q2FsaWJyaSI+TmlyLDwvc3Bh bj48bzpwPjwvbzpwPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2lu LXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gc3R5bGU9ImZv bnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6Q2FsaWJyaSI+Jm5ic3A7PC9zcGFuPjxvOnA+PC9v OnA+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDph dXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBzdHlsZT0iZm9udC1zaXplOjEx LjBwdDtmb250LWZhbWlseTpDYWxpYnJpIj5Jc27igJl0IFNQTSBtYW5hZ2VkIHZpYSBTYW5sb2Nr PyBJIGJlbGlldmUgdGhlcmUgaXMgbm8gbmVlZCB0byBmZW5jZSBTUE0gaG9zdC4gRXNwZWNpYWxs eSBpZiB0aGVyZSBhcmUgbm8gU1BNIHRhc2tzIHJ1bm5pbmcuPC9zcGFuPjxvOnA+PC9vOnA+PC9w Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y bWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29O b3JtYWwiPkl0J3MgdHJ1ZSB0aGF0IHRoZSBleGNsdXNpdml0eSBvZiB0aGUgU1BNIHJvbGUgaXMg ZW5mb3JjZWQgYnkgU2FubG9jayBidXQgeW91IGFsd2F5cyBuZWVkIHRvIGZlbmNlIGEgbm9uLXJl c3BvbnNpdmUgU1BNIGJlY2F1c2UgdGhlcmUgaXMgbm8gd2F5IHRvIGd1YXJhbnRlZSB0aGF0IHRo ZSBob3N0IGlzIG5vdCBzdGlsbCBtYW5pcHVsYXRpbmcgc3RvcmFnZSAoZWcuIExWIGV4dGVuc2lv bnMpIGFuZCB3ZSBtdXN0DQogZW5zdXJlIHRoYXQgb25seSBvbmUgaG9zdCBoYXMgdGhlIG1hc3Rl cmZzIG9uIHRoZSBtYXN0ZXIgc3RvcmFnZSBkb21haW4gbW91bnRlZC48bzpwPjwvbzpwPjwvcD4N CjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPiZuYnNwOzxvOnA+PC9vOnA+PC9w Pg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29s aWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQu OHB0O21hcmdpbi1yaWdodDowY20iPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs IiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1 dG8iPjxzcGFuIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OkNhbGlicmkiPiZu YnNwOzwvc3Bhbj48bzpwPjwvbzpwPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJt c28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4g c3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6Q2FsaWJyaSI+Jm5ic3A7PC9zcGFu PjxvOnA+PC9vOnA+PC9wPg0KPGRpdiBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLXRvcDpzb2xp ZCAjQjVDNERGIDEuMHB0O3BhZGRpbmc6My4wcHQgMGNtIDBjbSAwY20iPg0KPHAgY2xhc3M9Ik1z b05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9t LWFsdDphdXRvIj48Yj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6Q2FsaWJyaTtjb2xvcjpibGFj ayI+RnJvbToNCjwvc3Bhbj48L2I+PHNwYW4gc3R5bGU9ImZvbnQtZmFtaWx5OkNhbGlicmk7Y29s b3I6YmxhY2siPiZsdDs8YSBocmVmPSJtYWlsdG86dXNlcnMtYm91bmNlc0BvdmlydC5vcmciIHRh cmdldD0iX2JsYW5rIj51c2Vycy1ib3VuY2VzQG92aXJ0Lm9yZzwvYT4mZ3Q7IG9uIGJlaGFsZiBv ZiBOaXIgU29mZmVyICZsdDs8YSBocmVmPSJtYWlsdG86bnNvZmZlckByZWRoYXQuY29tIiB0YXJn ZXQ9Il9ibGFuayI+bnNvZmZlckByZWRoYXQuY29tPC9hPiZndDs8YnI+DQo8Yj5EYXRlOiA8L2I+ TW9uZGF5LCAxNyBBcHJpbCAyMDE3IGF0IDE2OjA2PGJyPg0KPGI+VG86IDwvYj5Lb25zdGFudGlu IFJhc2tvc2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzprb25yYXNrb0BnbWFpbC5jb20iIHRhcmdl dD0iX2JsYW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+Jmd0OywgRGFuIFlhc255ICZsdDs8YSBo cmVmPSJtYWlsdG86ZHlhc255QGdtYWlsLmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmR5YXNueUBnbWFp bC5jb208L2E+Jmd0Ozxicj4NCjxiPkNjOiA8L2I+dXNlcnMgJmx0OzxhIGhyZWY9Im1haWx0bzp1 c2Vyc0BvdmlydC5vcmciIHRhcmdldD0iX2JsYW5rIj51c2Vyc0BvdmlydC5vcmc8L2E+Jmd0Oywg RkVSTkFORE8gRlJFRElBTkkgJmx0OzxhIGhyZWY9Im1haWx0bzpmZXJuYW5kby5mcmVkaWFuaUB1 cHguY29tIiB0YXJnZXQ9Il9ibGFuayI+ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbTwvYT4mZ3Q7 PGJyPg0KPGI+U3ViamVjdDogPC9iPlJlOiBbb3ZpcnQtdXNlcnNdIHN0b3JhZ2UgcmVkdW5kYW5j eSBpbiBPdmlydDwvc3Bhbj48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNz PSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJv dHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2 Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6 YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+T24gTW9uLCBBcHIgMTcsIDIwMTcgYXQg ODoyNCBBTSBLb25zdGFudGluIFJhc2tvc2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzprb25yYXNr b0BnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+Jmd0OyB3 cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpu b25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2 LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tdG9wOjUuMHB0O21hcmdpbi1yaWdodDowY207 bWFyZ2luLWJvdHRvbTo1LjBwdCI+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9 Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5CdXQg YWN0dWFsbHksIGl0IGRpZG4ndCB3b3JrIHdlbGwuIEFmdGVyIG1haW4gU1BNIGhvc3Qgd2VudCBk b3duIEkgc2VlIHRoaXMNCjxvOnA+PC9vOnA+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6 YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNv Tm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20t YWx0OmF1dG8iPjxvOnA+Jm5ic3A7PC9vOnA+PC9wPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0i TXNvTm9ybWFsIj4yMDE3LTA0LTE3IDA1OjIzOjE1LDU1NFogRVJST1IgW29yZy5vdmlydC5lbmdp bmUuY29yZS52ZHNicm9rZXIuaXJzYnJva2VyLklyc1Byb3h5XSAoRGVmYXVsdFF1YXJ0elNjaGVk dWxlcjUpIFs0ZGNjMDMzZC0yNmJmLTQ5YmItYmZhYS0wM2E5NzBkYmJlYzFdIFNQTSBJbml0OiBj b3VsZCBub3QgZmluZCByZXBvcnRlZCB2ZHMgb3Igbm90IHVwIC0gcG9vbDogJ1NURycgdmRzX3Nw bV9pZDogJzEnPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0K PGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3At YWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjIwMTctMDQtMTcgMDU6MjM6MTUs NTY3WiBJTkZPICZuYnNwO1tvcmcub3ZpcnQuZW5naW5lLmNvcmUudmRzYnJva2VyLmlyc2Jyb2tl ci5JcnNQcm94eV0gKERlZmF1bHRRdWFydHpTY2hlZHVsZXI1KSBbNGRjYzAzM2QtMjZiZi00OWJi LWJmYWEtMDNhOTcwZGJiZWMxXSBTUE0gc2VsZWN0aW9uIC0gdmRzIHNlZW1zIGFzDQogc3BtICd0 YW5rNSc8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi IHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0 byI+MjAxNy0wNC0xNyAwNToyMzoxNSw1NjdaIFdBUk4gJm5ic3A7W29yZy5vdmlydC5lbmdpbmUu Y29yZS52ZHNicm9rZXIuaXJzYnJva2VyLklyc1Byb3h5XSAoRGVmYXVsdFF1YXJ0elNjaGVkdWxl cjUpIFs0ZGNjMDMzZC0yNmJmLTQ5YmItYmZhYS0wM2E5NzBkYmJlYzFdIHNwbSB2ZHMgaXMgbm9u IHJlc3BvbnNpdmUsIHN0b3BwaW5nDQogc3BtIHNlbGVjdGlvbi48bzpwPjwvbzpwPjwvcD4NCjwv ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h bHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+ DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10 b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPlNvIHRoYXQgbWVhbnMgb25s eSBpZiBCTUMgaXMgdXAgaXQncyBwb3NzaWJsZSB0byBhdXRvbWF0aWNhbGx5IHN3aXRjaCAmbmJz cDtTUE0gaG9zdD88bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2 Pg0KPC9ibG9ja3F1b3RlPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1h bCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDph dXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29O b3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1h bHQ6YXV0byI+Qk1DPzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1z b05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9t LWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNz PSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJv dHRvbS1hbHQ6YXV0byI+SWYgeW91ciBTUE0gaXMgbm8gcmVzcG9uc2l2ZSwgdGhlIHN5c3RlbSB3 aWxsIHRyeSB0byBmZW5jZSBpdC4gRGlkIHlvdSZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+ DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDph dXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5jb25maWd1cmUgcG93ZXIgbWFuYWdlbWVu dCBmb3IgYWxsIGhvc3RzPyBkaWQgeW91IGNoZWNrIHRoYXQgaXQmbmJzcDs8bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2lu LXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+d29yaz8gSG93IGRpZCB5 b3Ugc2ltdWxhdGUgbm9uLXJlc3BvbnNpdmUgaG9zdD8mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwv ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h bHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+ DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10 b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPklmIHBvd2VyIG1hbmFnZW1l bnQgaXMgbm90IGNvbmZpZ3VyZWQgb3IgZmFpbCwgdGhlIHN5c3RlbSBjYW5ub3Q8bzpwPjwvbzpw PjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+bW92ZSB0aGUgc3Bt IHRvIGFub3RoZXIgaG9zdCwgdW5sZXNzIHlvdSBtYW51YWxseSBjb25maXJtIHRoYXQgdGhlPG86 cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPlNQTSBo b3N0IHdhcyByZWJvb3RlZC48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNz PSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJv dHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp bi1ib3R0b20tYWx0OmF1dG8iPk5pcjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAg Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJn aW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGJsb2Nr cXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7 cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tdG9wOjUu MHB0O21hcmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1LjBwdCI+DQo8ZGl2Pg0KPGRpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxk aXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87 bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPlRoYW5rczxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+ DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10 b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+ PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h bHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+T24gU3VuLCBBcHIgMTYsIDIwMTcg YXQgODoyOSBQTSwgS29uc3RhbnRpbiBSYXNrb3NobnlpICZsdDs8YSBocmVmPSJtYWlsdG86a29u cmFza29AZ21haWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+a29ucmFza29AZ21haWwuY29tPC9hPiZn dDsgd3JvdGU6PG86cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7 Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0 O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTttYXJn aW4tYm90dG9tOjUuMHB0Ij4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNv LW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPk9oLCBmZW5j ZSBhZ2VudCB3b3JrcyBmaW5lIGlmIEkgc2VsZWN0IGlsbzQsPG86cD48L286cD48L3A+DQo8L2Rp dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0 OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPlRoYW5rIHlvdSBmb3IgeW91ciBoZWxw ITxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0i TXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0 b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFz cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i b3R0b20tYWx0OmF1dG8iPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDg6MjIgUE0gRGFuIFlhc255 ICZsdDs8YSBocmVmPSJtYWlsdG86ZHlhc255QGdtYWlsLmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmR5 YXNueUBnbWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGJs b2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4w cHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tdG9w OjUuMHB0O21hcmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1LjBwdCI+DQo8ZGl2Pg0KPGRp dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0 OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0 IDExOjE5IFBNLCBLb25zdGFudGluIFJhc2tvc2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzprb25y YXNrb0BnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+Jmd0 OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTti b3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7 bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdp bi1ib3R0b206NS4wcHQiPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t bWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+TWFrZXMgc2Vu c2UuJm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y bWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0 OmF1dG8iPkkgd2FzIHRyeWluZyB0byBzZXQgaXQgdXAsIGJ1dCBkb2Vzbid0IHdvcmsgd2l0aCBv dXIgc3RhZ2luZyBoYXJkd2FyZS4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+V2UgaGF2ZSBvbGQgaWxvMTAwLCBJJ2xsIHRyeSBhZ2Fp bi4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6 YXV0byI+VGhhbmtzIQ0KPG86cD48L286cD48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRv bS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+ DQo8L2Jsb2NrcXVvdGU+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1t YXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpw PjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+ DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRv cC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+SXQgaXMgYWJzb2x1dGVseSBu ZWNlc3NhcnkgZm9yIGFueSBIQSB0byB3b3JrIHByb3Blcmx5LiBUaGVyZSdzIG9mIGNvdXJzZSB0 aGUgJnF1b3Q7Y29uZmlybSBob3N0IGhhcyBiZWVuIHNodXRkb3duJnF1b3Q7IG9wdGlvbiwgd2hp Y2ggc2VydmVzIGFzIGFuIG92ZXJyaWRlIGZvciB0aGUgZmVuY2UgY29tbWFuZCwgYnV0IGl0J3MN CiBtYW51YWw8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0K PGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJt c28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7 PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTti b3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7 bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdp bi1ib3R0b206NS4wcHQiPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp bi1ib3R0b20tYWx0OmF1dG8iPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDg6MTggUE0gRGFuIFlh c255ICZsdDs8YSBocmVmPSJtYWlsdG86ZHlhc255QGdtYWlsLmNvbSIgdGFyZ2V0PSJfYmxhbmsi PmR5YXNueUBnbWFpbC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0K PGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0Mg MS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4t dG9wOjUuMHB0O21hcmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1LjBwdCI+DQo8ZGl2Pg0K PGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3At YWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPk9uIFN1biwgQXByIDE2LCAyMDE3 IGF0IDExOjE1IFBNLCBLb25zdGFudGluIFJhc2tvc2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzpr b25yYXNrb0BnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+ Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwvcD4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9u ZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4w cHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21h cmdpbi1ib3R0b206NS4wcHQiPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJt c28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+RmVuY2Ug YWdlbnQgdW5kZXIgZWFjaCBub2RlPzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Jsb2NrcXVv dGU+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs dDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4N CjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRp dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+V2hlbiB5b3UgY29uZmlndXJlIGEgaG9zdCwgdGhl cmUncyB0aGUgcG93ZXIgbWFuYWdlbWVudCB0YWIsIHdoZXJlIHlvdSBuZWVkIHRvIGVudGVyIHRo ZSBibWMgZGV0YWlscyBmb3IgdGhlIGhvc3QuIElmIHlvdSBkb24ndCBoYXZlIGZlbmNpbmcgZW5h YmxlZCwgaG93IGRvIHlvdSBleHBlY3QgdGhlIHN5c3RlbQ0KIHRvIG1ha2Ugc3VyZSBhIGhvc3Qg cnVubmluZyBhIHNlcnZpY2UgaXMgYWN0dWFsbHkgZG93biAoYW5kIGl0IGlzIHNhZmUgdG8gc3Rh cnQgSEEgc2VydmljZXMgZWxzZXdoZXJlKSwgYW5kIG5vdCwgZm9yIGV4YW1wbGUsIGp1c3QgdW5y ZWFjaGFibGUgYnkgdGhlIGVuZ2luZT8gSG93IGRvIHlvdSBhdm9pZCBhIHNwbGl0YnJhaWQgLSZn dDsgU0JBID88bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0K PGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJt c28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7 PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTti b3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7 bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdp bi1ib3R0b206NS4wcHQiPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1h bCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDph dXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs dDphdXRvIj5PbiBTdW4sIEFwciAxNiwgMjAxNyBhdCA4OjE0IFBNIERhbiBZYXNueSAmbHQ7PGEg aHJlZj0ibWFpbHRvOmR5YXNueUBnbWFpbC5jb20iIHRhcmdldD0iX2JsYW5rIj5keWFzbnlAZ21h aWwuY29tPC9hPiZndDsgd3JvdGU6PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3Rl IHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRp bmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDtt YXJnaW4tcmlnaHQ6MGNtO21hcmdpbi1ib3R0b206NS4wcHQiPg0KPGRpdj4NCjxkaXY+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21z by1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5PbiBTdW4sIEFwciAxNiwgMjAxNyBhdCAxMToxMyBQ TSwgS29uc3RhbnRpbiBSYXNrb3NobnlpICZsdDs8YSBocmVmPSJtYWlsdG86a29ucmFza29AZ21h aWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+a29ucmFza29AZ21haWwuY29tPC9hPiZndDsgd3JvdGU6 PG86cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxl ZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1s ZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTttYXJnaW4tYm90dG9t OjUuMHB0Ij4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10 b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZxdW90O0Nvcm5lciBjYXNl cyZxdW90Oz8mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRv bS1hbHQ6YXV0byI+SSB0cmllZCB0byBzaW11bGF0ZSBjcmFzaCBvZiBTUE0gc2VydmVyIGFuZCBv dmlydCBrZXB0IHRyeWluZyB0byByZWlzdGFibGlzaGVkIGNvbm5lY3Rpb24gdG8gdGhlIGZhaWxl ZCBub2RlLiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21z by1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0K PC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNs YXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2lu LWJvdHRvbS1hbHQ6YXV0byI+RGlkIHlvdSBjb25maWd1cmUgZmVuY2luZz88bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0 bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rp dj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0ND Q0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFy Z2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdpbi1ib3R0b206NS4wcHQiPg0KPGRp dj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t dG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpw PjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48 L286cD48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t bWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+T24gU3VuLCBB cHIgMTYsIDIwMTcgYXQgODoxMCBQTSBEYW4gWWFzbnkgJmx0OzxhIGhyZWY9Im1haWx0bzpkeWFz bnlAZ21haWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+ZHlhc255QGdtYWlsLmNvbTwvYT4mZ3Q7IHdy b3RlOjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5v bmU7Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYu MHB0O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTtt YXJnaW4tYm90dG9tOjUuMHB0Ij4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29O b3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1h bHQ6YXV0byI+T24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgNzoyOSBBTSwgTmlyIFNvZmZlciAmbHQ7 PGEgaHJlZj0ibWFpbHRvOm5zb2ZmZXJAcmVkaGF0LmNvbSIgdGFyZ2V0PSJfYmxhbmsiPm5zb2Zm ZXJAcmVkaGF0LmNvbTwvYT4mZ3Q7IHdyb3RlOjxvOnA+PC9vOnA+PC9wPg0KPGJsb2NrcXVvdGUg c3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGlu ZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJnaW4tdG9wOjUuMHB0O21h cmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1LjBwdCI+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+ DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNv LW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPk9uIFN1biwgQXByIDE2LCAyMDE3IGF0IDI6MDUgUE0g RGFuIFlhc255ICZsdDs8YSBocmVmPSJtYWlsdG86ZHlhc255QHJlZGhhdC5jb20iIHRhcmdldD0i X2JsYW5rIj5keWFzbnlAcmVkaGF0LmNvbTwvYT4mZ3Q7IHdyb3RlOjxvOnA+PC9vOnA+PC9wPg0K PC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29saWQg I0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0 O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTttYXJnaW4tYm90dG9tOjUuMHB0Ij4N CjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwv cD4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0 OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0 bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+T24gQXByIDE2LCAyMDE3IDc6MDEgQU0sICZx dW90O05pciBTb2ZmZXImcXVvdDsgJmx0OzxhIGhyZWY9Im1haWx0bzpuc29mZmVyQHJlZGhhdC5j b20iIHRhcmdldD0iX2JsYW5rIj5uc29mZmVyQHJlZGhhdC5jb208L2E+Jmd0OyB3cm90ZTo8bzpw PjwvbzpwPjwvcD4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpz b2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6 NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdpbi1ib3R0b206NS4w cHQiPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0 eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+ T24gU3VuLCBBcHIgMTYsIDIwMTcgYXQgNDoxNyBBTSBEYW4gWWFzbnkgJmx0OzxhIGhyZWY9Im1h aWx0bzpkeWFzbnlAZ21haWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+ZHlhc255QGdtYWlsLmNvbTwv YT4mZ3Q7IHdyb3RlOjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0i Ym9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAw Y20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJp Z2h0OjBjbTttYXJnaW4tYm90dG9tOjUuMHB0Ij4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs IiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1 dG8iPldoZW4geW91IHNldCB1cCBhIHN0b3JhZ2UgZG9tYWluLCB5b3UgbmVlZCB0byBzcGVjaWZ5 IGEgaG9zdCB0byBwZXJmb3JtIHRoZSBpbml0aWFsIHN0b3JhZ2Ugb3BlcmF0aW9ucywgYnV0IG9u Y2UgdGhlIFNEIGlzIGRlZmluZWQsIGl0J3MgZGV0YWlscyBhcmUgaW4gdGhlIGVuZ2luZSBkYXRh YmFzZSwgYW5kDQogYWxsIHRoZSBob3N0cyBnZXQgY29ubmVjdGVkIHRvIGl0IGRpcmVjdGx5LiBJ ZiB0aGUgZmlyc3QgaG9zdCB5b3UgdXNlZCB0byBkZWZpbmUgdGhlIFNEIGdvZXMgZG93biwgYWxs IG90aGVyIGhvc3RzIHdpbGwgc3RpbGwgcmVtYWluIGNvbm5lY3RlZCBhbmQgd29yay4gU1BNIGlz IGFuIEhBIHNlcnZpY2UsIGFuZCBpZiB0aGUgY3VycmVudCBTUE0gaG9zdCBnb2VzIGRvd24sIFNQ TSBnZXRzIHN0YXJ0ZWQgb24gYW5vdGhlciBob3N0IGluIHRoZSBEQy4NCiBJbiBzaG9ydCwgdW5s ZXNzIHlvdXIgYWN0dWFsIE5GUyBleHBvcnRpbmcgaG9zdCBnb2VzIGRvd24sIHRoZXJlIGlzIG5v IG91dGFnZS4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPGRp dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4N CjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRv cC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+VGhlcmUgaXMgbm8gc3RvcmFn ZSBvdXRhZ2UsIGJ1dCBpZiB5b3Ugc2h1dGRvd24gdGhlIHNwbSBob3N0LCB0aGUgc3BtIGhvc3Q8 bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+d2ls bCBub3QgbW92ZSB0byBhIG5ldyBob3N0IHVudGlsIHRoZSBzcG0gaG9zdCBpcyBvbmxpbmUgYWdh aW4sIG9yIHlvdSBjb25maXJtPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFz cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i b3R0b20tYWx0OmF1dG8iPm1hbnVhbGx5IHRoYXQgdGhlIHNwbSBob3N0IHdhcyByZWJvb3RlZC48 bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwv ZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNw OzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90 dG9tLWFsdDphdXRvIj5JbiBhIHByb3Blcmx5IGNvbmZpZ3VyZWQgc2V0dXAgdGhlIFNCQSBzaG91 bGQgdGFrZSBjYXJlIG9mIHRoYXQuIFRoYXQncyB0aGUgd2hvbGUgcG9pbnQgb2YgSEEgc2Vydmlj ZXMmbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8 ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRv O21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2 Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6 YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+SW4gc29tZSBjYXNlcyBsaWtlIHBvd2Vy IGxvc3Mgb3IgaGFyZHdhcmUgZmFpbHVyZSwgdGhlcmUgaXMgbm8gd2F5IHRvIHN0YXJ0PG86cD48 L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNv LW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPnRoZSBzcG0g aG9zdCwgYW5kIHRoZSBzeXN0ZW0gY2Fubm90IHJlY292ZXIgYXV0b21hdGljYWxseS48bzpwPjwv bzpwPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjxkaXY+DQo8 cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1h cmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rp dj4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9 Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90 dG9tLWFsdDphdXRvIj5UaGVyZSBhcmUgYWx3YXlzIGNvcm5lciBjYXNlcywgbm8gZG91YnQuIEJ1 dCBpbiBhIG5vcm1hbCBzaXR1YXRpb24uIHdoZXJlIGFuIFNQTSBob3N0IGdvZXMgZG93biBiZWNh dXNlIG9mIGEgaGFyZHdhcmUgZmFpbHVyZSwgaXQgZ2V0cyBmZW5jZWQsIG90aGVyIGhvc3RzIGNv bnRlbmQgZm9yIFNQTSBhbmQgc3RhcnQNCiBpdC4gTm8gc3VycHJpc2VzIHRoZXJlLiZuYnNwOzxv OnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRp dj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t dG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpw PjwvcD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0 OnNvbGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVm dDo0LjhwdDttYXJnaW4tdG9wOjUuMHB0O21hcmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1 LjBwdCI+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFu IHN0eWxlPSJjb2xvcjojODg4ODg4Ij4mbmJzcDs8L3NwYW4+PG86cD48L286cD48L3A+DQo8L2Rp dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0 OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIHN0eWxlPSJjb2xvcjojODg4 ODg4Ij5OaXI8L3NwYW4+PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRp dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4N CjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0ND IDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2lu LXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdpbi1ib3R0b206NS4wcHQiPg0KPGRpdj4N CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1 dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9k aXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7 Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0 O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTttYXJn aW4tYm90dG9tOjUuMHB0Ij4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6 YXV0byI+PHNwYW4gc3R5bGU9ImNvbG9yOiM4ODg4ODgiPiZuYnNwOzwvc3Bhbj48bzpwPjwvbzpw PjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gc3R5bGU9 ImNvbG9yOiM4ODg4ODgiPk5pcjwvc3Bhbj48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4N CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1 dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9k aXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0ND Q0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21h cmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTttYXJnaW4tYm90dG9tOjUuMHB0Ij4NCjxk aXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87 bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPGRpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+T24gU2F0LCBBcHIgMTUsIDIwMTcgYXQgMTo1MyBQTSwg S29uc3RhbnRpbiBSYXNrb3NobnlpICZsdDs8YSBocmVmPSJtYWlsdG86a29ucmFza29AZ21haWwu Y29tIiB0YXJnZXQ9Il9ibGFuayI+a29ucmFza29AZ21haWwuY29tPC9hPiZndDsgd3JvdGU6PG86 cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6 c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0 OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBjbTttYXJnaW4tYm90dG9tOjUu MHB0Ij4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3At YWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPkhpIEZlcm5hbmRvLDxvOnA+PC9v OnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1t YXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5JIHNlZSBlYWNo IGhvc3QgaGFzIGRpcmVjdCBjb25uZWN0aW9uIG5mcyBtb3VudCwgYnV0IHllcywgaWYgbWFpbiBo b3N0IHRvIHdoaWNoIEkgY29ubmVjdGVkIG5mcyBzdG9yYWdlIGdvaW5nIGRvd24gdGhlIHN0b3Jh Z2UgYmVjb21lcyB1bmF2YWlsYWJsZSBhbmQgYWxsIHZtcyBhcmUgZG93bjxvOnA+PC9vOnA+PC9w Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZu YnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIg c3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRv Ij4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1h bCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDph dXRvIj5PbiBTYXQsIEFwciAxNSwgMjAxNyBhdCAxMDozNyBBTSBGRVJOQU5ETyBGUkVESUFOSSAm bHQ7PGEgaHJlZj0ibWFpbHRvOmZlcm5hbmRvLmZyZWRpYW5pQHVweC5jb20iIHRhcmdldD0iX2Js YW5rIj5mZXJuYW5kby5mcmVkaWFuaUB1cHguY29tPC9hPiZndDsgd3JvdGU6PG86cD48L286cD48 L3A+DQo8L2Rpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpz b2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAwY20gNi4wcHQ7bWFyZ2luLWxlZnQ6 NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6MGNtO21hcmdpbi1ib3R0b206NS4w cHQiPg0KPGRpdj4NCjxkaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0 eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttYXJnaW4tYm90dG9tOjEyLjBwdCI+SGVsbG8g S29uc3RhbnRpbi48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIg c3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRv Ij5UaGF0IGRvZXNuYHQgbWFrZSBtdWNoIHNlbnNlIG1ha2UgYSB3aG9sZSBjbHVzdGVyIGRlcGVu ZCBvbiBhIHNpbmdsZSBob3N0LiBGcm9tIHdoYXQgSSBrbm93IGFueSBob3N0IHRhbGsgZGlyZWN0 bHkgdG8gTkZTIFN0b3JhZ2UgQXJyYXkgb3Igd2hhdGV2ZXIgb3RoZXIgU2hhcmVkIFN0b3JhZ2Ug eW91IGhhdmUuPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0 eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttYXJnaW4tYm90dG9tOjEyLjBwdCI+SGF2ZSB5 b3UgdGVzdGVkIHRoYXQgaG9zdCBnb2luZyBkb3duIGlmIHRoYXQgYWZmZWN0cyB0aGUgb3RoZXIg d2l0aCB0aGUgTkZTIG1vdW50ZWQgZGlyZWN0bGx5IGluIGEgTkZTIFN0b3JhZ2UgYXJyYXkgPzxv OnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs IiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1 dG8iPkZlcm5hbmRvPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNv Tm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20t YWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6 YXV0byI+MjAxNy0wNC0xNSAxMjo0MiBHTVQtMDM6MDAgS29uc3RhbnRpbiBSYXNrb3NobnlpICZs dDs8YSBocmVmPSJtYWlsdG86a29ucmFza29AZ21haWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+a29u cmFza29AZ21haWwuY29tPC9hPiZndDs6PG86cD48L286cD48L3A+DQo8YmxvY2txdW90ZSBzdHls ZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBj bSAwY20gMGNtIDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2lu LXJpZ2h0OjBjbTttYXJnaW4tYm90dG9tOjUuMHB0Ij4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y bWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0 OmF1dG8iPkluIG92aXJ0IHlvdSBoYXZlIHRvIGF0dGFjaCBzdG9yYWdlIHRocm91Z2ggc3BlY2lm aWMgaG9zdC4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRv bS1hbHQ6YXV0byI+SWYgaG9zdCBnb2VzIGRvd24gc3RvcmFnZSBpcyBub3QgYXZhaWxhYmxlLiZu YnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIg c3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRv Ij4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1h bCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDph dXRvIj5PbiBTYXQsIEFwciAxNSwgMjAxNyBhdCA3OjMxIEFNIEZFUk5BTkRPIEZSRURJQU5JICZs dDs8YSBocmVmPSJtYWlsdG86ZmVybmFuZG8uZnJlZGlhbmlAdXB4LmNvbSIgdGFyZ2V0PSJfYmxh bmsiPmZlcm5hbmRvLmZyZWRpYW5pQHVweC5jb208L2E+Jmd0OyB3cm90ZTo8bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNv bGlkICNDQ0NDQ0MgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0 LjhwdDttYXJnaW4tdG9wOjUuMHB0O21hcmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1LjBw dCI+DQo8ZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNv LW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPldlbGwsIG1h a2UgaXQgbm90IGdvIHRocm91Z2ggaG9zdDEgYW5kIGRlZGljYXRlIGEgc3RvcmFnZSBzZXJ2ZXIg Zm9yIHJ1bm5pbmcgTkZTIGFuZCBtYWtlIGJvdGggaG9zdHMgY29ubmVjdCB0byBpdC48bzpwPjwv bzpwPjwvcD4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t dG9wLWFsdDphdXRvO21hcmdpbi1ib3R0b206MTIuMHB0Ij5JbiBteSB2aWV3IE5GUyBpcyBtdWNo IGVhc2llciB0byBtYW5hZ2UgdGhhbiBhbnkgb3RoZXIgdHlwZSBvZiBzdG9yYWdlLCBzcGVjaWFs bHkgRkMgYW5kIGlTQ1NJIGFuZCBwZXJmb3JtYW5jZSBpcyBwcmV0dHkgbXVjaCB0aGUgc2FtZSwg c28geW91IHdvbmB0IGdldCBiZXR0ZXIgcmVzdWx0cyBvdGhlciB0aGFuIG1hbmFnZW1lbnQNCiBn b2luZyB0byBvdGhlciB0eXBlLjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNv Tm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20t YWx0OmF1dG8iPkZlcm5hbmRvPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFz cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i b3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPGRp dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+MjAxNy0wNC0xNSA1OjI1IEdNVC0wMzowMCBLb25z dGFudGluIFJhc2tvc2hueWkgJmx0OzxhIGhyZWY9Im1haWx0bzprb25yYXNrb0BnbWFpbC5jb20i IHRhcmdldD0iX2JsYW5rIj5rb25yYXNrb0BnbWFpbC5jb208L2E+Jmd0Ozo8bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxibG9ja3F1b3RlIHN0eWxlPSJib3Jk ZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCAjQ0NDQ0NDIDEuMHB0O3BhZGRpbmc6MGNtIDBjbSAw Y20gNi4wcHQ7bWFyZ2luLWxlZnQ6NC44cHQ7bWFyZ2luLXRvcDo1LjBwdDttYXJnaW4tcmlnaHQ6 MGNtO21hcmdpbi1ib3R0b206NS4wcHQiPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0 eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+ SGkgZ3V5cywNCjxvOnA+PC9vOnA+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0 eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+ SSBoYXZlIG9uZSBuZnMmbmJzcDtzdG9yYWdlLDxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21z by1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5pdCdzIGNvbm5lY3RlZCB0aHJvdWdoIGhvc3QxLjxv OnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9 Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5ob3N0 MiBhbHNvIGhhcyBhY2Nlc3MgdG8gaXQsIEkgY2FuIGVhc2lseSBtaWdyYXRlIHZtcyZuYnNwO2Jl dHdlZW4gdGhlbS48bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29O b3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1h bHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0i TXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0 b20tYWx0OmF1dG8iPlRoZSBxdWVzdGlvbiBpcyAtIGlmIGhvc3QxIGlzIGRvd24gLSBhbGwgaW5m cmFzdHJ1Y3R1cmUgaXMgZG93biwgc2luY2UgYWxsIHRyYWZmaWMgZ29lcyB0aHJvdWdoIGhvc3Qx LDxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5 bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj5p cyB0aGVyZSBhbnkgd2F5IGluIG9WaXJ0IHRvIHVzZSByZWR1bmRhbnQgc3RvcmFnZT88bzpwPjwv bzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t bWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86 cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPk9ubHkg Z2x1c3RlcmZzPzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs dDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRv bS1hbHQ6YXV0byI+VGhhbmtzPG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFz cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i b3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjxw IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFy Z2luLWJvdHRvbS1hbHQ6YXV0byI+Jm5ic3A7PG86cD48L286cD48L3A+DQo8L2Jsb2NrcXVvdGU+ DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8YmxvY2txdW90ZSBzdHlsZT0iYm9yZGVy Om5vbmU7Ym9yZGVyLWxlZnQ6c29saWQgI0NDQ0NDQyAxLjBwdDtwYWRkaW5nOjBjbSAwY20gMGNt IDYuMHB0O21hcmdpbi1sZWZ0OjQuOHB0O21hcmdpbi10b3A6NS4wcHQ7bWFyZ2luLXJpZ2h0OjBj bTttYXJnaW4tYm90dG9tOjUuMHB0Ij4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t bWFyZ2luLXRvcC1hbHQ6YXV0bzttYXJnaW4tYm90dG9tOjEyLjBwdCI+X19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX188YnI+DQpVc2VycyBtYWlsaW5nIGxpc3Q8 YnI+DQo8YSBocmVmPSJtYWlsdG86VXNlcnNAb3ZpcnQub3JnIiB0YXJnZXQ9Il9ibGFuayI+VXNl cnNAb3ZpcnQub3JnPC9hPjxicj4NCjxhIGhyZWY9Imh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFp bG1hbi9saXN0aW5mby91c2VycyIgdGFyZ2V0PSJfYmxhbmsiPmh0dHA6Ly9saXN0cy5vdmlydC5v cmcvbWFpbG1hbi9saXN0aW5mby91c2VyczwvYT48bzpwPjwvbzpwPjwvcD4NCjwvYmxvY2txdW90 ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ibG9j a3F1b3RlPg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10 b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+ PC9wPg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rp dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt YXJnaW4tYm90dG9tOjEyLjBwdCI+PGJyPg0KX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX188YnI+DQpVc2VycyBtYWlsaW5nIGxpc3Q8YnI+DQo8YSBocmVmPSJt YWlsdG86VXNlcnNAb3ZpcnQub3JnIiB0YXJnZXQ9Il9ibGFuayI+VXNlcnNAb3ZpcnQub3JnPC9h Pjxicj4NCjxhIGhyZWY9Imh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0aW5mby91 c2VycyIgdGFyZ2V0PSJfYmxhbmsiPmh0dHA6Ly9saXN0cy5vdmlydC5vcmcvbWFpbG1hbi9saXN0 aW5mby91c2VyczwvYT48bzpwPjwvbzpwPjwvcD4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPHAg Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJn aW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpwPjwvbzpwPjwvcD4NCjwvZGl2Pg0KPHAgY2xh c3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4t Ym90dG9tLWFsdDphdXRvIj5fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fXzxicj4NClVzZXJzIG1haWxpbmcgbGlzdDxicj4NCjxhIGhyZWY9Im1haWx0bzpVc2Vy c0BvdmlydC5vcmciIHRhcmdldD0iX2JsYW5rIj5Vc2Vyc0BvdmlydC5vcmc8L2E+PGJyPg0KPGEg aHJlZj0iaHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL3VzZXJzIiB0YXJn ZXQ9Il9ibGFuayI+aHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL3VzZXJz PC9hPjxvOnA+PC9vOnA+PC9wPg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2 Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21h cmdpbi1ib3R0b206MTIuMHB0Ij48YnI+DQpfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fXzxicj4NClVzZXJzIG1haWxpbmcgbGlzdDxicj4NCjxhIGhyZWY9Im1h aWx0bzpVc2Vyc0BvdmlydC5vcmciIHRhcmdldD0iX2JsYW5rIj5Vc2Vyc0BvdmlydC5vcmc8L2E+ PGJyPg0KPGEgaHJlZj0iaHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3RpbmZvL3Vz ZXJzIiB0YXJnZXQ9Il9ibGFuayI+aHR0cDovL2xpc3RzLm92aXJ0Lm9yZy9tYWlsbWFuL2xpc3Rp bmZvL3VzZXJzPC9hPjxvOnA+PC9vOnA+PC9wPg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp bi1ib3R0b20tYWx0OmF1dG8iPiZuYnNwOzxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8L2Rpdj4N CjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8 L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rp dj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+ DQo8L2Rpdj4NCjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0K PC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPC9k aXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2 Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4N CjwvYmxvY2txdW90ZT4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1t YXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj4mbmJzcDs8bzpw PjwvbzpwPjwvcD4NCjwvZGl2Pg0KPC9ibG9ja3F1b3RlPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2 Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1h cmdpbi1ib3R0b206MTIuMHB0Ij48YnI+DQpfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fXzxicj4NClVzZXJzIG1haWxpbmcgbGlzdDxicj4NCjxhIGhyZWY9Im1h aWx0bzpVc2Vyc0BvdmlydC5vcmciPlVzZXJzQG92aXJ0Lm9yZzwvYT48YnI+DQo8YSBocmVmPSJo dHRwOi8vbGlzdHMub3ZpcnQub3JnL21haWxtYW4vbGlzdGluZm8vdXNlcnMiIHRhcmdldD0iX2Js YW5rIj5odHRwOi8vbGlzdHMub3ZpcnQub3JnL21haWxtYW4vbGlzdGluZm8vdXNlcnM8L2E+PG86 cD48L286cD48L3A+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi Pjxicj4NCjxiciBjbGVhcj0iYWxsIj4NCjxicj4NCi0tIDxvOnA+PC9vOnA+PC9wPg0KPGRpdj4N CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj5BZGFtIExpdGtlPG86cD48L286cD48L3A+DQo8 L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ib2R5Pg0KPC9odG1sPg0K --_000_87040B69038146B985204DA5C6360EEAacroniscom_--
On Mon, Apr 17, 2017 at 11:04 AM, Pavel Gashev <Pax@acronis.com> wrote:
Adam,
You know, Sanlock has recovery mechanism that kills VDSM, or even triggers Watchdog to reboot SPM host in case it has lost the SPM lock.
I’m asking because I had issues with my master storage that caused SPM host to reboot by Watchdog. And I was sure that it’s an intended behaviour. Isn’t it?
Yes of course. But an SPM host can fail but still maintain its connection to the storage lease. In this case still you need classic fencing. Something new we are investigating is the use of sanlock's request feature which allows a new host to take the lease away from the current holder. The current holder would be fenced by sanlock (watchdog if necessary) and only once the lease is free would it be granted to the new requester.
*From: *Adam Litke <alitke@redhat.com> *Date: *Monday, 17 April 2017 at 17:32 *To: *Pavel Gashev <Pax@acronis.com> *Cc: *Nir Soffer <nsoffer@redhat.com>, users <users@ovirt.org>
*Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 9:26 AM, Pavel Gashev <Pax@acronis.com> wrote:
Nir,
Isn’t SPM managed via Sanlock? I believe there is no need to fence SPM host. Especially if there are no SPM tasks running.
It's true that the exclusivity of the SPM role is enforced by Sanlock but you always need to fence a non-responsive SPM because there is no way to guarantee that the host is not still manipulating storage (eg. LV extensions) and we must ensure that only one host has the masterfs on the master storage domain mounted.
*From: *<users-bounces@ovirt.org> on behalf of Nir Soffer < nsoffer@redhat.com> *Date: *Monday, 17 April 2017 at 16:06 *To: *Konstantin Raskoshnyi <konrasko@gmail.com>, Dan Yasny < dyasny@gmail.com> *Cc: *users <users@ovirt.org>, FERNANDO FREDIANI < fernando.frediani@upx.com> *Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1'
2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5'
2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you
configure power management for all hosts? did you check that it
work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot
move the spm to another host, unless you manually confirm that the
SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4,
Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense.
I was trying to set it up, but doesn't work with our staging hardware.
We have old ilo100, I'll try again.
Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"?
I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host
will not move to a new host until the spm host is online again, or you confirm
manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start
the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
Hi Fernando,
I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.
Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host.
If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.
In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys,
I have one nfs storage,
it's connected through host1.
host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,
is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--
Adam Litke
-- Adam Litke
On Mon, Apr 17, 2017 at 6:54 PM Adam Litke <alitke@redhat.com> wrote:
On Mon, Apr 17, 2017 at 11:04 AM, Pavel Gashev <Pax@acronis.com> wrote:
Adam,
You know, Sanlock has recovery mechanism that kills VDSM, or even triggers Watchdog to reboot SPM host in case it has lost the SPM lock.
I’m asking because I had issues with my master storage that caused SPM host to reboot by Watchdog. And I was sure that it’s an intended behaviour. Isn’t it?
Yes of course. But an SPM host can fail but still maintain its connection to the storage lease. In this case still you need classic fencing.
Something new we are investigating is the use of sanlock's request feature which allows a new host to take the lease away from the current holder. The current holder would be fenced by sanlock (watchdog if necessary) and only once the lease is free would it be granted to the new requester.
We can use the SPM lease to kill vdsm on the non-responsive SPM host, and start the SPM on another host, similar to the way we handle vms with a lease. But this does not help with the masterfs mounted on the SPM host. if vdsm is killed before it unmount it, strating the SPM on another host (and mounting the msasterfs on the new host) will corrupt the masterfs. When using file based storage (nfs, glusterfs) we don't have a masterfs so killing vdsm on the SPM should be good enough to start the SPM on another host, even if fencing is not possible. We can start with enabling sanlock based SPM fencing on file based storage. Nir
*From: *Adam Litke <alitke@redhat.com> *Date: *Monday, 17 April 2017 at 17:32 *To: *Pavel Gashev <Pax@acronis.com> *Cc: *Nir Soffer <nsoffer@redhat.com>, users <users@ovirt.org>
*Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 9:26 AM, Pavel Gashev <Pax@acronis.com> wrote:
Nir,
Isn’t SPM managed via Sanlock? I believe there is no need to fence SPM host. Especially if there are no SPM tasks running.
It's true that the exclusivity of the SPM role is enforced by Sanlock but you always need to fence a non-responsive SPM because there is no way to guarantee that the host is not still manipulating storage (eg. LV extensions) and we must ensure that only one host has the masterfs on the master storage domain mounted.
*From: *<users-bounces@ovirt.org> on behalf of Nir Soffer < nsoffer@redhat.com> *Date: *Monday, 17 April 2017 at 16:06 *To: *Konstantin Raskoshnyi <konrasko@gmail.com>, Dan Yasny < dyasny@gmail.com> *Cc: *users <users@ovirt.org>, FERNANDO FREDIANI < fernando.frediani@upx.com> *Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1'
2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5'
2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you
configure power management for all hosts? did you check that it
work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot
move the spm to another host, unless you manually confirm that the
SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4,
Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense.
I was trying to set it up, but doesn't work with our staging hardware.
We have old ilo100, I'll try again.
Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"?
I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host
will not move to a new host until the spm host is online again, or you confirm
manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start
the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando,
I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.
Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host.
If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.
In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys,
I have one nfs storage,
it's connected through host1.
host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,
is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--
Adam Litke
-- Adam Litke
--_000_3162c49395404987a301406953a54eb4acroniscom_ Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Nir, A process can chdir into mount point and then lazy umount it. Filesystem re= mains mounted while the process exists and current directory is on mounted = filesystem. # struncate -s 1g masterfs.img # mkfs.ext4 masterfs.img # mkdir masterfs # mount -o loop masterfs.img masterfs # cd masterfs # umount -l . # touch file # ls # cd .. # ls masterfs ________________________________ From: Nir Soffer <nsoffer@redhat.com> Sent: Apr 17, 2017 8:40 PM To: Adam Litke; Pavel Gashev Cc: users Subject: Re: [ovirt-users] storage redundancy in Ovirt On Mon, Apr 17, 2017 at 6:54 PM Adam Litke <alitke@redhat.com<mailto:alitke= @redhat.com>> wrote: On Mon, Apr 17, 2017 at 11:04 AM, Pavel Gashev <Pax@acronis.com<mailto:Pax@= acronis.com>> wrote: Adam, You know, Sanlock has recovery mechanism that kills VDSM, or even triggers = Watchdog to reboot SPM host in case it has lost the SPM lock. I=92m asking because I had issues with my master storage that caused SPM ho= st to reboot by Watchdog. And I was sure that it=92s an intended behaviour.= Isn=92t it? Yes of course. But an SPM host can fail but still maintain its connection = to the storage lease. In this case still you need classic fencing. Something new we are investigating is the use of sanlock's request feature = which allows a new host to take the lease away from the current holder. Th= e current holder would be fenced by sanlock (watchdog if necessary) and onl= y once the lease is free would it be granted to the new requester. We can use the SPM lease to kill vdsm on the non-responsive SPM host, and start the SPM on another host, similar to the way we handle vms with a lease. But this does not help with the masterfs mounted on the SPM host. if vdsm is killed before it unmount it, strating the SPM on another host (and mount= ing the msasterfs on the new host) will corrupt the masterfs. When using file based storage (nfs, glusterfs) we don't have a masterfs so killing vdsm on the SPM should be good enough to start the SPM on another host, even if fencing is not possible. We can start with enabling sanlock based SPM fencing on file based storage. Nir From: Adam Litke <alitke@redhat.com<mailto:alitke@redhat.com>> Date: Monday, 17 April 2017 at 17:32 To: Pavel Gashev <Pax@acronis.com<mailto:Pax@acronis.com>> Cc: Nir Soffer <nsoffer@redhat.com<mailto:nsoffer@redhat.com>>, users <user= s@ovirt.org<mailto:users@ovirt.org>> Subject: Re: [ovirt-users] storage redundancy in Ovirt On Mon, Apr 17, 2017 at 9:26 AM, Pavel Gashev <Pax@acronis.com<mailto:Pax@a= cronis.com>> wrote: Nir, Isn=92t SPM managed via Sanlock? I believe there is no need to fence SPM ho= st. Especially if there are no SPM tasks running. It's true that the exclusivity of the SPM role is enforced by Sanlock but y= ou always need to fence a non-responsive SPM because there is no way to gua= rantee that the host is not still manipulating storage (eg. LV extensions) = and we must ensure that only one host has the masterfs on the master storag= e domain mounted. From: <users-bounces@ovirt.org<mailto:users-bounces@ovirt.org>> on behalf o= f Nir Soffer <nsoffer@redhat.com<mailto:nsoffer@redhat.com>> Date: Monday, 17 April 2017 at 16:06 To: Konstantin Raskoshnyi <konrasko@gmail.com<mailto:konrasko@gmail.com>>, = Dan Yasny <dyasny@gmail.com<mailto:dyasny@gmail.com>> Cc: users <users@ovirt.org<mailto:users@ovirt.org>>, FERNANDO FREDIANI <fer= nando.frediani@upx.com<mailto:fernando.frediani@upx.com>> Subject: Re: [ovirt-users] storage redundancy in Ovirt On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com<m= ailto:konrasko@gmail.com>> wrote: But actually, it didn't work well. After main SPM host went down I see this 2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.I= rsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] S= PM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1= ' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.I= rsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] S= PM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.I= rsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] s= pm vds is non responsive, stopping spm selection. So that means only if BMC is up it's possible to automatically switch SPM = host? BMC? If your SPM is no responsive, the system will try to fence it. Did you configure power management for all hosts? did you check that it work? How did you simulate non-responsive host? If power management is not configured or fail, the system cannot move the spm to another host, unless you manually confirm that the SPM host was rebooted. Nir Thanks On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi <konrasko@gmail.com<= mailto:konrasko@gmail.com>> wrote: Oh, fence agent works fine if I select ilo4, Thank you for your help! On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com<mailto:dyasny@g= mail.com>> wrote: On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi <konrasko@gmail.com= <mailto:konrasko@gmail.com>> wrote: Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks! It is absolutely necessary for any HA to work properly. There's of course t= he "confirm host has been shutdown" option, which serves as an override for= the fence command, but it's manual On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com<mailto:dyasny@g= mail.com>> wrote: On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi <konrasko@gmail.com= <mailto:konrasko@gmail.com>> wrote: Fence agent under each node? When you configure a host, there's the power management tab, where you need= to enter the bmc details for the host. If you don't have fencing enabled, = how do you expect the system to make sure a host running a service is actua= lly down (and it is safe to start HA services elsewhere), and not, for exam= ple, just unreachable by the engine? How do you avoid a splitbraid -> SBA ? On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com<mailto:dyasny@g= mail.com>> wrote: On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi <konrasko@gmail.com= <mailto:konrasko@gmail.com>> wrote: "Corner cases"? I tried to simulate crash of SPM server and ovirt kept trying to reistablis= hed connection to the failed node. Did you configure fencing? On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com<mailto:dyasny@g= mail.com>> wrote: On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com<mailto:nsof= fer@redhat.com>> wrote: On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com<mailto:dyasny@= redhat.com>> wrote: On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com<mailto:nsoffer@re= dhat.com>> wrote: On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com<mailto:dyasny@g= mail.com>> wrote: When you set up a storage domain, you need to specify a host to perform the= initial storage operations, but once the SD is defined, it's details are i= n the engine database, and all the hosts get connected to it directly. If t= he first host you used to define the SD goes down, all other hosts will sti= ll remain connected and work. SPM is an HA service, and if the current SPM = host goes down, SPM gets started on another host in the DC. In short, unles= s your actual NFS exporting host goes down, there is no outage. There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you conf= irm manually that the spm host was rebooted. In a properly configured setup the SBA should take care of that. That's the= whole point of HA services In some cases like power loss or hardware failure, there is no way to start the spm host, and the system cannot recover automatically. There are always corner cases, no doubt. But in a normal situation. where a= n SPM host goes down because of a hardware failure, it gets fenced, other h= osts contend for SPM and start it. No surprises there. Nir Nir On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi <konrasko@gmail.com<= mailto:konrasko@gmail.com>> wrote: Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to w= hich I connected nfs storage going down the storage becomes unavailable and= all vms are down On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <fernando.frediani@upx.c= om<mailto:fernando.frediani@upx.com>> wrote: Hello Konstantin. That doesn`t make much sense make a whole cluster depend on a single host. =
From what I know any host talk directly to NFS Storage Array or whatever ot= her Shared Storage you have. Have you tested that host going down if that affects the other with the NFS= mounted directlly in a NFS Storage array ? Fernando
></span> wrote:<br> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex; border-left:1=
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com<mailto= :konrasko@gmail.com>>: In ovirt you have to attach storage through specific host. If host goes down storage is not available. On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <fernando.frediani@upx.co= m<mailto:fernando.frediani@upx.com>> wrote: Well, make it not go through host1 and dedicate a storage server for runnin= g NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, spe= cially FC and iSCSI and performance is pretty much the same, so you won`t g= et better results other than management going to other type. Fernando 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com<mailto:= konrasko@gmail.com>>: Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them. The question is - if host1 is down - all infrastructure is down, since all = traffic goes through host1, is there any way in oVirt to use redundant storage? Only glusterfs? Thanks _______________________________________________ Users mailing list Users@ovirt.org<mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users _______________________________________________ Users mailing list Users@ovirt.org<mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users _______________________________________________ Users mailing list Users@ovirt.org<mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users _______________________________________________ Users mailing list Users@ovirt.org<mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users _______________________________________________ Users mailing list Users@ovirt.org<mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users -- Adam Litke -- Adam Litke --_000_3162c49395404987a301406953a54eb4acroniscom_ Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable <html> <head> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1= 252"> <meta content=3D"text/html; charset=3Dutf-8"> </head> <body> <style type=3D"text/css"> <!-- div#d_1492464244266 p {font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12.0pt; color:#1F497D} --> </style> <div id=3D"d_1492464244266" style=3D"font-family:Calibri,Arial,Helvetica,sa= ns-serif; font-size:12.0pt; color:#1F497D"> <p dir=3D"ltr">Nir,</p> <p dir=3D"ltr">A process can chdir into mount point and then lazy umount it= . Filesystem remains mounted while the process exists and current directory= is on mounted filesystem.</p> <p dir=3D"ltr"># struncate -s 1g masterfs.img<br> # mkfs.ext4 masterfs.img<br> # mkdir masterfs<br> # mount -o loop masterfs.img masterfs<br> # cd masterfs<br> # umount -l .<br> # touch file<br> # ls<br> # cd ..<br> # ls masterfs</p> <div id=3D"signature-x" style=3D"font-family:Calibri,Arial,Helvetica,sans-s= erif; font-size:12.0pt; color:#1F497D"> <br> </div> </div> <div id=3D"quoted_header" style=3D"clear:both"> <hr style=3D"border:none; height:1px; color:#E1E1E1; background-color:#E1E1= E1"> <div style=3D"border:none; padding:3.0pt 0cm 0cm 0cm"><span style=3D"font-s= ize:11.0pt; font-family:'Calibri','sans-serif'"><b>From:</b> Nir Soffer <= ;nsoffer@redhat.com><br> <b>Sent:</b> Apr 17, 2017 8:40 PM<br> <b>To:</b> Adam Litke; Pavel Gashev<br> <b>Cc:</b> users<br> <b>Subject:</b> Re: [ovirt-users] storage redundancy in Ovirt<br> </span></div> </div> <br type=3D"attribution"> <div> <div dir=3D"ltr"> <div class=3D"gmail_quote"> <div dir=3D"ltr">On Mon, Apr 17, 2017 at 6:54 PM Adam Litke <<a href=3D"= mailto:alitke@redhat.com">alitke@redhat.com</a>> wrote:<br> </div> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex; border-left:1= px #ccc solid; padding-left:1ex"> <div dir=3D"ltr"> <div class=3D"gmail_extra"> <div class=3D"gmail_quote">On Mon, Apr 17, 2017 at 11:04 AM, Pavel Gashev <= span dir=3D"ltr"> <<a href=3D"mailto:Pax@acronis.com" target=3D"_blank">Pax@acronis.com</a= px #ccc solid; padding-left:1ex"> <div bgcolor=3D"white" lang=3D"EN-GB"> <div class=3D"m_1785069678384749853m_-3628883287347250028WordSection1"> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= ">Adam,<u></u><u></u></span></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "><u></u> <u></u></span></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= ">You know, Sanlock has recovery mechanism that kills VDSM, or even trigger= s Watchdog to reboot SPM host in case it has lost the SPM lock.<u></u><u></= u></span></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= ">I=92m asking because I had issues with my master storage that caused SPM = host to reboot by Watchdog. And I was sure that it=92s an intended behaviou= r. Isn=92t it?</span></p> </div> </div> </blockquote> <div><br> </div> </div> </div> </div> <div dir=3D"ltr"> <div class=3D"gmail_extra"> <div class=3D"gmail_quote"> <div>Yes of course. But an SPM host can fail but still maintain its c= onnection to the storage lease. In this case still you need classic f= encing.<br> <br> Something new we are investigating is the use of sanlock's request feature = which allows a new host to take the lease away from the current holder.&nbs= p; The current holder would be fenced by sanlock (watchdog if necessary) an= d only once the lease is free would it be granted to the new requester. <br> </div> </div> </div> </div> </blockquote> <div><br> </div> <div>We can use the SPM lease to kill vdsm on the non-responsive SPM host,<= /div> <div>and start the SPM on another host, similar to the way we handle vms wi= th</div> <div>a lease.</div> <div><br> </div> <div>But this does not help with the masterfs mounted on the SPM host. if v= dsm</div> <div>is killed before it unmount it, strating the SPM on another host (and = mounting</div> <div>the msasterfs on the new host) will corrupt the masterfs.</div> <div><br> </div> <div>When using file based storage (nfs, glusterfs) we don't have a masterf= s so</div> <div>killing vdsm on the SPM should be good enough to start the SPM on anot= her</div> <div>host, even if fencing is not possible.</div> <div><br> </div> <div>We can start with enabling sanlock based SPM fencing on file based sto= rage.</div> <div><br> </div> <div>Nir</div> <div> </div> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex; border-left:1= px #ccc solid; padding-left:1ex"> <div dir=3D"ltr"> <div class=3D"gmail_extra"> <div class=3D"gmail_quote"> <div></div> </div> </div> </div> <div dir=3D"ltr"> <div class=3D"gmail_extra"> <div class=3D"gmail_quote"> <div> </div> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex; border-left:1= px #ccc solid; padding-left:1ex"> <div bgcolor=3D"white" lang=3D"EN-GB"> <div class=3D"m_1785069678384749853m_-3628883287347250028WordSection1"> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "><u></u><u></u></span></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "><u></u> <u></u></span></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "><u></u> <u></u></span></p> <div style=3D"border:none; border-top:solid #b5c4df 1.0pt; padding:3.0pt 0c= m 0cm 0cm"> <p class=3D"MsoNormal"><b><span style=3D"font-family:Calibri; color:black">= From: </span> </b><span style=3D"font-family:Calibri; color:black">Adam Litke <<a href= =3D"mailto:alitke@redhat.com" target=3D"_blank">alitke@redhat.com</a>><b= r> <b>Date: </b>Monday, 17 April 2017 at 17:32<br> <b>To: </b>Pavel Gashev <<a href=3D"mailto:Pax@acronis.com" target=3D"_b= lank">Pax@acronis.com</a>><br> <b>Cc: </b>Nir Soffer <<a href=3D"mailto:nsoffer@redhat.com" target=3D"_= blank">nsoffer@redhat.com</a>>, users <<a href=3D"mailto:users@ovirt.= org" target=3D"_blank">users@ovirt.org</a>></span></p> <div> <div class=3D"m_1785069678384749853h5"><br> <b>Subject: </b>Re: [ovirt-users] storage redundancy in Ovirt<u></u><u></u>= </div> </div> <p></p> </div> <div> <div class=3D"m_1785069678384749853h5"> <div> <p class=3D"MsoNormal"><u></u> <u></u></p> </div> <div> <p class=3D"MsoNormal"><u></u> <u></u></p> <div> <p class=3D"MsoNormal"><u></u> <u></u></p> <div> <p class=3D"MsoNormal">On Mon, Apr 17, 2017 at 9:26 AM, Pavel Gashev <<a= href=3D"mailto:Pax@acronis.com" target=3D"_blank">Pax@acronis.com</a>> = wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-right:0cm"> <div> <div> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= ">Nir,</span><u></u><u></u></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "> </span><u></u><u></u></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= ">Isn=92t SPM managed via Sanlock? I believe there is no need to fence SPM = host. Especially if there are no SPM tasks running.</span><u></u><u></u></p=
</u></p> </div> <div> <p class=3D"MsoNormal"><span style=3D"color:#888888">Nir</span><u></u><u></= u></p> </div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <div> <div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal"><span style=3D"color:#888888"> </span><u></u><u= </u></p> </div> <div> <p class=3D"MsoNormal"><span style=3D"color:#888888">Nir</span><u></u><u></= u></p> </div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <p class=3D"MsoNormal">On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshn= yi <<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gma= il.com</a>> wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">Hi Fernando,<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">I see each host has direct connection nfs mount, but= yes, if main host to which I connected nfs storage going down the storage = becomes unavailable and all vms are down<u></u><u></u></p> </div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal">On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI &= lt;<a href=3D"mailto:fernando.frediani@upx.com" target=3D"_blank">fernando.= frediani@upx.com</a>> wrote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <div> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt">Hello Konstantin.<u><= /u><u></u></p> </div> <p class=3D"MsoNormal">That doesn`t make much sense make a whole cluster de=
</div> </div> </blockquote> <div> <p class=3D"MsoNormal"><u></u> <u></u></p> </div> <div> <p class=3D"MsoNormal">It's true that the exclusivity of the SPM role is en= forced by Sanlock but you always need to fence a non-responsive SPM because= there is no way to guarantee that the host is not still manipulating stora= ge (eg. LV extensions) and we must ensure that only one host has the masterfs on the master storage domain mo= unted.<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-right:0cm"> <div> <div> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "> </span><u></u><u></u></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt; font-family:Calibri= "> </span><u></u><u></u></p> <div style=3D"border:none; border-top:solid #b5c4df 1.0pt; padding:3.0pt 0c= m 0cm 0cm"> <p class=3D"MsoNormal"><b><span style=3D"font-family:Calibri; color:black">= From: </span> </b><span style=3D"font-family:Calibri; color:black"><<a href=3D"mailto:= users-bounces@ovirt.org" target=3D"_blank">users-bounces@ovirt.org</a>> = on behalf of Nir Soffer <<a href=3D"mailto:nsoffer@redhat.com" target=3D= "_blank">nsoffer@redhat.com</a>><br> <b>Date: </b>Monday, 17 April 2017 at 16:06<br> <b>To: </b>Konstantin Raskoshnyi <<a href=3D"mailto:konrasko@gmail.com" = target=3D"_blank">konrasko@gmail.com</a>>, Dan Yasny <<a href=3D"mail= to:dyasny@gmail.com" target=3D"_blank">dyasny@gmail.com</a>><br> <b>Cc: </b>users <<a href=3D"mailto:users@ovirt.org" target=3D"_blank">u= sers@ovirt.org</a>>, FERNANDO FREDIANI <<a href=3D"mailto:fernando.fr= ediani@upx.com" target=3D"_blank">fernando.frediani@upx.com</a>><br> <b>Subject: </b>Re: [ovirt-users] storage redundancy in Ovirt</span><u></u>= <u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <div> <div> <p class=3D"MsoNormal">On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshny= i <<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gmai= l.com</a>> wrote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">But actually, it didn't work well. After main SPM ho= st went down I see this <u></u><u></u></p> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"><u></u> <u></u></p> <div> <div> <p class=3D"MsoNormal">2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.cor= e.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49= bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - poo= l: 'STG' vds_spm_id: '1'<u></u><u></u></p> </div> </div> </div> <div> <div> <div> <p class=3D"MsoNormal">2017-04-17 05:23:15,567Z INFO [org.ovirt.engin= e.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26= bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5'<u></u><= u></u></p> </div> <div> <p class=3D"MsoNormal">2017-04-17 05:23:15,567Z WARN [org.ovirt.engin= e.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26= bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selectio= n.<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">So that means only if BMC is up it's possible to aut= omatically switch SPM host?<u></u><u></u></p> </div> </div> </div> </div> </blockquote> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">BMC?<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">If your SPM is no responsive, the system will try to= fence it. Did you <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">configure power management for all hosts? did you ch= eck that it <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">work? How did you simulate non-responsive host? = ;<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">If power management is not configured or fail, the s= ystem cannot<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">move the spm to another host, unless you manually co= nfirm that the<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">SPM host was rebooted.<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">Nir<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">Thanks<u></u><u></u></p> </div> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshn= yi <<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gma= il.com</a>> wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">Oh, fence agent works fine if I select ilo4,<u></u><= u></u></p> </div> <div> <p class=3D"MsoNormal">Thank you for your help!<u></u><u></u></p> </div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <<a hre= f=3D"mailto:dyasny@gmail.com" target=3D"_blank">dyasny@gmail.com</a>> wr= ote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskosh= nyi <<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gm= ail.com</a>> wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">Makes sense. <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">I was trying to set it up, but doesn't work with our= staging hardware. <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">We have old ilo100, I'll try again. <u></u><u><= /u></p> </div> <div> <p class=3D"MsoNormal">Thanks! <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> </div> </blockquote> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal">It is absolutely necessary for any HA to work proper= ly. There's of course the "confirm host has been shutdown" option= , which serves as an override for the fence command, but it's manual<u></u>= <u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <<a hre= f=3D"mailto:dyasny@gmail.com" target=3D"_blank">dyasny@gmail.com</a>> wr= ote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskosh= nyi <<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gm= ail.com</a>> wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">Fence agent under each node?<u></u><u></u></p> </div> </blockquote> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal">When you configure a host, there's the power managem= ent tab, where you need to enter the bmc details for the host. If you don't= have fencing enabled, how do you expect the system to make sure a host run= ning a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unr= eachable by the engine? How do you avoid a splitbraid -> SBA ?<u></u><u>= </u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <<a hre= f=3D"mailto:dyasny@gmail.com" target=3D"_blank">dyasny@gmail.com</a>> wr= ote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskosh= nyi <<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gm= ail.com</a>> wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">"Corner cases"? <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">I tried to simulate crash of SPM server and ovirt ke= pt trying to reistablished connection to the failed node. <u></u><u></= u></p> </div> </blockquote> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal">Did you configure fencing?<u></u><u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <<a hre= f=3D"mailto:dyasny@gmail.com" target=3D"_blank">dyasny@gmail.com</a>> wr= ote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <<a h= ref=3D"mailto:nsoffer@redhat.com" target=3D"_blank">nsoffer@redhat.com</a>&= gt; wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <<a hre= f=3D"mailto:dyasny@redhat.com" target=3D"_blank">dyasny@redhat.com</a>> = wrote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <p class=3D"MsoNormal">On Apr 16, 2017 7:01 AM, "Nir Soffer" <= <a href=3D"mailto:nsoffer@redhat.com" target=3D"_blank">nsoffer@redhat.com<= /a>> wrote:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <div> <p class=3D"MsoNormal">On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <<a hre= f=3D"mailto:dyasny@gmail.com" target=3D"_blank">dyasny@gmail.com</a>> wr= ote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">When you set up a storage domain, you need to specif= y a host to perform the initial storage operations, but once the SD is defi= ned, it's details are in the engine database, and all the hosts get connect= ed to it directly. If the first host you used to define the SD goes down, all other hosts will still remain con= nected and work. SPM is an HA service, and if the current SPM host goes dow= n, SPM gets started on another host in the DC. In short, unless your actual= NFS exporting host goes down, there is no outage. <u></u><u></u></p> </div> </blockquote> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> <div> <p class=3D"MsoNormal">There is no storage outage, but if you shutdown the = spm host, the spm host<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">will not move to a new host until the spm host is on= line again, or you confirm<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">manually that the spm host was rebooted.<u></u><u></= u></p> </div> </div> </div> </blockquote> </div> </div> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> <div> <div> <p class=3D"MsoNormal">In a properly configured setup the SBA should take c= are of that. That's the whole point of HA services <u></u><u></u></p> </div> </div> </blockquote> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">In some cases like power loss or hardware failure, t= here is no way to start<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">the spm host, and the system cannot recover automati= cally.<u></u><u></u></p> </div> </div> </div> </blockquote> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal">There are always corner cases, no doubt. But in a no= rmal situation. where an SPM host goes down because of a hardware failure, = it gets fenced, other hosts contend for SPM and start it. No surprises ther= e. <u></u><u></u></p> </div> </div> </div> </div> <div> <div> <div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal"><span style=3D"color:#888888"> </span><u></u><u= pend on a single host. From what I know any host talk directly to NFS Stora= ge Array or whatever other Shared Storage you have.<u></u><u></u></p> </div> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt">Have you tested that = host going down if that affects the other with the NFS mounted directlly in= a NFS Storage array ?<u></u><u></u></p> </div> </div> <div> <p class=3D"MsoNormal">Fernando<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <p class=3D"MsoNormal">2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <= ;<a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gmail.com= </a>>:<u></u><u></u></p> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">In ovirt you have to attach storage through specific= host. <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">If host goes down storage is not available. <u>= </u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal">On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI &l= t;<a href=3D"mailto:fernando.frediani@upx.com" target=3D"_blank">fernando.f= rediani@upx.com</a>> wrote:<u></u><u></u></p> </div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <div> <div> <p class=3D"MsoNormal">Well, make it not go through host1 and dedicate a st= orage server for running NFS and make both hosts connect to it.<u></u><u></= u></p> </div> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt">In my view NFS is muc= h easier to manage than any other type of storage, specially FC and iSCSI a= nd performance is pretty much the same, so you won`t get better results oth= er than management going to other type.<u></u><u></u></p> </div> <p class=3D"MsoNormal">Fernando<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <div> <p class=3D"MsoNormal">2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <= <a href=3D"mailto:konrasko@gmail.com" target=3D"_blank">konrasko@gmail.com<= /a>>:<u></u><u></u></p> </div> </div> <div> <div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <div> <p class=3D"MsoNormal">Hi guys, <u></u><u></u></p> <div> <p class=3D"MsoNormal">I have one nfs storage,<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">it's connected through host1.<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">host2 also has access to it, I can easily migrate vm= s between them.<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">The question is - if host1 is down - all infrastruct= ure is down, since all traffic goes through host1,<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">is there any way in oVirt to use redundant storage?<= u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">Only glusterfs?<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <div> <p class=3D"MsoNormal">Thanks<u></u><u></u></p> </div> <div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> <p class=3D"MsoNormal"> <u></u><u></u></p> </blockquote> </div> </div> <div> <div> <blockquote style=3D"border:none; border-left:solid #cccccc 1.0pt; padding:= 0cm 0cm 0cm 6.0pt; margin-left:4.8pt; margin-top:5.0pt; margin-right:0cm; m= argin-bottom:5.0pt"> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt">_____________________= __________________________<br> Users mailing list<br> <a href=3D"mailto:Users@ovirt.org" target=3D"_blank">Users@ovirt.org</a><br=
<a href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank"=
http://lists.ovirt.org/mailman/listinfo/users</a><u></u><u></u></p> </blockquote> </div> </div> </blockquote> </div> </div> </blockquote> </div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </blockquote> </div> </div> </div> </div> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><br> _______________________________________________<br> Users mailing list<br> <a href=3D"mailto:Users@ovirt.org" target=3D"_blank">Users@ovirt.org</a><br=
<a href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank"=
http://lists.ovirt.org/mailman/listinfo/users</a><u></u><u></u></p> </blockquote> </div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> <p class=3D"MsoNormal">_______________________________________________<br> Users mailing list<br> <a href=3D"mailto:Users@ovirt.org" target=3D"_blank">Users@ovirt.org</a><br=
<a href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank"=
http://lists.ovirt.org/mailman/listinfo/users</a><u></u><u></u></p> </blockquote> </div> </div> </div> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><br> _______________________________________________<br> Users mailing list<br> <a href=3D"mailto:Users@ovirt.org" target=3D"_blank">Users@ovirt.org</a><br=
<a href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank"=
http://lists.ovirt.org/mailman/listinfo/users</a><u></u><u></u></p> </blockquote> </div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> </div> </div> </blockquote> </div> </div> </div> </div> </blockquote> </div> <p class=3D"MsoNormal"> <u></u><u></u></p> </div> </blockquote> </div> </div> </div> </div> </div> </div> <p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><br> _______________________________________________<br> Users mailing list<br> <a href=3D"mailto:Users@ovirt.org" target=3D"_blank">Users@ovirt.org</a><br=
<a href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank"=
http://lists.ovirt.org/mailman/listinfo/users</a><u></u><u></u></p> </blockquote> </div> <p class=3D"MsoNormal"><br> <br clear=3D"all"> <br> -- <u></u><u></u></p> <div> <div> <p class=3D"MsoNormal">Adam Litke<u></u><u></u></p> </div> </div> </div> </div> </div> </div> </div> </div> </blockquote> </div> </div> </div> <div dir=3D"ltr"> <div class=3D"gmail_extra"><br> <br clear=3D"all"> <br> -- <br> <div class=3D"m_1785069678384749853gmail_signature"> <div dir=3D"ltr">Adam Litke</div> </div> </div> </div> </blockquote> </div> </div> </div> </body> </html>
--_000_3162c49395404987a301406953a54eb4acroniscom_--
On Tue, Apr 18, 2017 at 12:23 AM Pavel Gashev <Pax@acronis.com> wrote:
Nir,
A process can chdir into mount point and then lazy umount it. Filesystem remains mounted while the process exists and current directory is on mounted filesystem.
# struncate -s 1g masterfs.img # mkfs.ext4 masterfs.img # mkdir masterfs # mount -o loop masterfs.img masterfs # cd masterfs # umount -l . # touch file # ls # cd .. # ls masterfs
Interesting idea! The only issue I see if not having a way to tell if the file system was actually unmounted. Does process termination guarantee that the file system was unmounted? Do you know if the behaviour is documented somewhere? Nir
------------------------------ *From:* Nir Soffer <nsoffer@redhat.com> *Sent:* Apr 17, 2017 8:40 PM *To:* Adam Litke; Pavel Gashev *Cc:* users
*Subject:* Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 6:54 PM Adam Litke <alitke@redhat.com> wrote:
On Mon, Apr 17, 2017 at 11:04 AM, Pavel Gashev <Pax@acronis.com> wrote:
Adam,
You know, Sanlock has recovery mechanism that kills VDSM, or even triggers Watchdog to reboot SPM host in case it has lost the SPM lock.
I’m asking because I had issues with my master storage that caused SPM host to reboot by Watchdog. And I was sure that it’s an intended behaviour. Isn’t it?
Yes of course. But an SPM host can fail but still maintain its connection to the storage lease. In this case still you need classic fencing.
Something new we are investigating is the use of sanlock's request feature which allows a new host to take the lease away from the current holder. The current holder would be fenced by sanlock (watchdog if necessary) and only once the lease is free would it be granted to the new requester.
We can use the SPM lease to kill vdsm on the non-responsive SPM host, and start the SPM on another host, similar to the way we handle vms with a lease.
But this does not help with the masterfs mounted on the SPM host. if vdsm is killed before it unmount it, strating the SPM on another host (and mounting the msasterfs on the new host) will corrupt the masterfs.
When using file based storage (nfs, glusterfs) we don't have a masterfs so killing vdsm on the SPM should be good enough to start the SPM on another host, even if fencing is not possible.
We can start with enabling sanlock based SPM fencing on file based storage.
Nir
*From: *Adam Litke <alitke@redhat.com> *Date: *Monday, 17 April 2017 at 17:32 *To: *Pavel Gashev <Pax@acronis.com> *Cc: *Nir Soffer <nsoffer@redhat.com>, users <users@ovirt.org>
*Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 9:26 AM, Pavel Gashev <Pax@acronis.com> wrote:
Nir,
Isn’t SPM managed via Sanlock? I believe there is no need to fence SPM host. Especially if there are no SPM tasks running.
It's true that the exclusivity of the SPM role is enforced by Sanlock but you always need to fence a non-responsive SPM because there is no way to guarantee that the host is not still manipulating storage (eg. LV extensions) and we must ensure that only one host has the masterfs on the master storage domain mounted.
*From: *<users-bounces@ovirt.org> on behalf of Nir Soffer < nsoffer@redhat.com> *Date: *Monday, 17 April 2017 at 16:06 *To: *Konstantin Raskoshnyi <konrasko@gmail.com>, Dan Yasny < dyasny@gmail.com> *Cc: *users <users@ovirt.org>, FERNANDO FREDIANI < fernando.frediani@upx.com> *Subject: *Re: [ovirt-users] storage redundancy in Ovirt
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1'
2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5'
2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you
configure power management for all hosts? did you check that it
work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot
move the spm to another host, unless you manually confirm that the
SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4,
Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense.
I was trying to set it up, but doesn't work with our staging hardware.
We have old ilo100, I'll try again.
Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Fence agent under each node?
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
"Corner cases"?
I tried to simulate crash of SPM server and ovirt kept trying to reistablished connection to the failed node.
Did you configure fencing?
On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host
will not move to a new host until the spm host is online again, or you confirm
manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
In some cases like power loss or hardware failure, there is no way to start
the spm host, and the system cannot recover automatically.
There are always corner cases, no doubt. But in a normal situation. where an SPM host goes down because of a hardware failure, it gets fenced, other hosts contend for SPM and start it. No surprises there.
Nir
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando,
I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.
Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host.
If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.
In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
Hi guys,
I have one nfs storage,
it's connected through host1.
host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,
is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--
Adam Litke
-- Adam Litke
--_000_14925401941105315camelacroniscom_ Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 T24gVHVlLCAyMDE3LTA0LTE4IGF0IDE3OjE5ICswMDAwLCBOaXIgU29mZmVyIHdyb3RlOg0KT24g VHVlLCBBcHIgMTgsIDIwMTcgYXQgMTI6MjMgQU0gUGF2ZWwgR2FzaGV2IDxQYXhAYWNyb25pcy5j b208bWFpbHRvOlBheEBhY3JvbmlzLmNvbT4+IHdyb3RlOg0KDQpOaXIsDQoNCkEgcHJvY2VzcyBj YW4gY2hkaXIgaW50byBtb3VudCBwb2ludCBhbmQgdGhlbiBsYXp5IHVtb3VudCBpdC4gRmlsZXN5 c3RlbSByZW1haW5zIG1vdW50ZWQgd2hpbGUgdGhlIHByb2Nlc3MgZXhpc3RzIGFuZCBjdXJyZW50 IGRpcmVjdG9yeSBpcyBvbiBtb3VudGVkIGZpbGVzeXN0ZW0uDQoNCiMgc3RydW5jYXRlIC1zIDFn IG1hc3RlcmZzLmltZw0KIyBta2ZzLmV4dDQgbWFzdGVyZnMuaW1nDQojIG1rZGlyIG1hc3RlcmZz DQojIG1vdW50IC1vIGxvb3AgbWFzdGVyZnMuaW1nIG1hc3RlcmZzDQojIGNkIG1hc3RlcmZzDQoj IHVtb3VudCAtbCAuDQojIHRvdWNoIGZpbGUNCiMgbHMNCiMgY2QgLi4NCiMgbHMgbWFzdGVyZnMN Cg0KSW50ZXJlc3RpbmcgaWRlYSENCg0KVGhlIG9ubHkgaXNzdWUgSSBzZWUgaWYgbm90IGhhdmlu ZyBhIHdheSB0byB0ZWxsIGlmIHRoZSBmaWxlIHN5c3RlbSB3YXMgYWN0dWFsbHkNCnVubW91bnRl ZC4gRG9lcyBwcm9jZXNzIHRlcm1pbmF0aW9uIGd1YXJhbnRlZSB0aGF0IHRoZSBmaWxlIHN5c3Rl bSB3YXMNCnVubW91bnRlZD8NCg0KWW91IGNhbiBkb3VibGUgY2hlY2sgdGhhdCBieSBkZWFjdGl2 YXRpbmcgdW5kZXJseWluZyBMVi4gSXQgc2hvdWxkIG5vdCBiZSBidXN5Lg0KDQpBcyBJIHVuZGVy c3RhbmQgaXQgdW5tb3VudHMgRlMgYXMgc29vbiBhcyBhbGwgaW5vZGVzIGFyZSBjbG9zZWQuIFNp bmNlIGtpbGxpbmcgb2YgcHJvY2VzcyBpcyBjbG9zaW5nIGFsbCBvcGVuIGlub2RlcywgeW91IGNh biBiZSBzdXJlIHRoYXQgZmlsZSBzeXN0ZW0gaXMgdW5tb3VudGVkIGlmIHRoZXJlIGFyZSBubyBv dGhlciBwcm9jZXNzZXMuIE9mIGNvdXJzZSB0aGlzIGlzIGEgcGxhY2UgZm9yIHJhY2UgY29uZGl0 aW9ucy4gRm9yIGV4YW1wbGUsIHVubW91bnRpbmcgY2FuIGJlIHN0dWNrIHNvbWUgdGltZSBpZiB1 bmRlcmx5aW5nIHN0b3JhZ2UgaXMgdW5yZXNwb25zaXZlLiBNdWx0aXBhdGggc2hvdWxkIGRyb3Ag YWxsIElPcyB3aXRoIEVJTy4gVkRTTSBoYXMgdG8gd2FpdCBlbm91Z2ggdGltZSBhZnRlciBhY3F1 aXJpbmcgdGhlIGxvY2sgYmVmb3JlIG1vdW50aW5nIG1hc3RlcmZzLg0KDQpEbyB5b3Uga25vdyBp ZiB0aGUgYmVoYXZpb3VyIGlzIGRvY3VtZW50ZWQgc29tZXdoZXJlPw0KDQp1bW91bnQoOCk6DQoN Ci1sLCAtLWxhenkNCkxhenkgdW5tb3VudC4gIERldGFjaCB0aGUgZmlsZXN5c3RlbSBmcm9tIHRo ZSBmaWxlc3lzdGVtIGhpZXJhcmNoeSBub3csIGFuZCBjbGVhbnVwIGFsbCByZWZlcmVuY2VzIHRv IHRoZSBmaWxlc3lzdGVtIGFzIHNvb24gYXMgaXQgaXMgbm90IGJ1c3kgYW55bW9yZS4gIChSZXF1 aXJlcyBrZXJuZWwgMi40LjExIG9yIGxhdGVyLikNCg0KDQo= --_000_14925401941105315camelacroniscom_ Content-Type: text/html; charset="utf-8" Content-ID: <4AC96B6347714B4BB1DF97D7D6BB2F00@acronis.com> Content-Transfer-Encoding: base64 PGh0bWw+DQo8aGVhZD4NCjxtZXRhIGh0dHAtZXF1aXY9IkNvbnRlbnQtVHlwZSIgY29udGVudD0i dGV4dC9odG1sOyBjaGFyc2V0PXV0Zi04Ij4NCjwvaGVhZD4NCjxib2R5Pg0KPGRpdj5PbiBUdWUs IDIwMTctMDQtMTggYXQgMTc6MTkgJiM0MzswMDAwLCBOaXIgU29mZmVyIHdyb3RlOjwvZGl2Pg0K PGJsb2NrcXVvdGUgdHlwZT0iY2l0ZSI+DQo8ZGl2IGRpcj0ibHRyIj4NCjxkaXYgY2xhc3M9Imdt YWlsX3F1b3RlIj4NCjxkaXYgZGlyPSJsdHIiPk9uIFR1ZSwgQXByIDE4LCAyMDE3IGF0IDEyOjIz IEFNIFBhdmVsIEdhc2hldiAmbHQ7PGEgaHJlZj0ibWFpbHRvOlBheEBhY3JvbmlzLmNvbSI+UGF4 QGFjcm9uaXMuY29tPC9hPiZndDsgd3JvdGU6PGJyPg0KPC9kaXY+DQo8YmxvY2txdW90ZSB0eXBl PSJjaXRlIj4NCjxkaXY+DQo8ZGl2IGlkPSJtXy01MjU5ODQwODU1NTYxOTUzNzU2ZF8xNDkyNDY0 MjQ0MjY2IiBzdHlsZT0iZm9udC1mYW1pbHk6Q2FsaWJyaSxBcmlhbCxIZWx2ZXRpY2Esc2Fucy1z ZXJpZjtmb250LXNpemU6MTIuMHB0O2NvbG9yOiMxZjQ5N2QiPg0KPHAgZGlyPSJsdHIiPk5pciw8 L3A+DQo8cCBkaXI9Imx0ciI+QSBwcm9jZXNzIGNhbiBjaGRpciBpbnRvIG1vdW50IHBvaW50IGFu ZCB0aGVuIGxhenkgdW1vdW50IGl0LiBGaWxlc3lzdGVtIHJlbWFpbnMgbW91bnRlZCB3aGlsZSB0 aGUgcHJvY2VzcyBleGlzdHMgYW5kIGN1cnJlbnQgZGlyZWN0b3J5IGlzIG9uIG1vdW50ZWQgZmls ZXN5c3RlbS48L3A+DQo8cCBkaXI9Imx0ciI+IyBzdHJ1bmNhdGUgLXMgMWcgbWFzdGVyZnMuaW1n PGJyPg0KIyBta2ZzLmV4dDQgbWFzdGVyZnMuaW1nPGJyPg0KIyBta2RpciBtYXN0ZXJmczxicj4N CiMgbW91bnQgLW8gbG9vcCBtYXN0ZXJmcy5pbWcgbWFzdGVyZnM8YnI+DQojIGNkIG1hc3RlcmZz PGJyPg0KIyB1bW91bnQgLWwgLjxicj4NCiMgdG91Y2ggZmlsZTxicj4NCiMgbHM8YnI+DQojIGNk IC4uPGJyPg0KIyBscyBtYXN0ZXJmczwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8YnI+DQo8L2Jsb2Nr cXVvdGU+DQo8ZGl2PkludGVyZXN0aW5nIGlkZWEhPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0K PGRpdj5UaGUgb25seSBpc3N1ZSBJIHNlZSBpZiBub3QgaGF2aW5nIGEgd2F5IHRvIHRlbGwgaWYg dGhlIGZpbGUgc3lzdGVtIHdhcyBhY3R1YWxseTwvZGl2Pg0KPGRpdj51bm1vdW50ZWQuIERvZXMg cHJvY2VzcyB0ZXJtaW5hdGlvbiBndWFyYW50ZWUgdGhhdCB0aGUgZmlsZSBzeXN0ZW0gd2FzPC9k aXY+DQo8ZGl2PnVubW91bnRlZD8mbmJzcDs8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2Nr cXVvdGU+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5Zb3UgY2FuIGRvdWJsZSBjaGVjayB0aGF0 IGJ5IGRlYWN0aXZhdGluZyB1bmRlcmx5aW5nIExWLiBJdCBzaG91bGQgbm90IGJlIGJ1c3kuPC9k aXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj5BcyBJIHVuZGVyc3RhbmQgaXQgdW5tb3VudHMg RlMgYXMgc29vbiBhcyBhbGwgaW5vZGVzIGFyZSBjbG9zZWQuIFNpbmNlIGtpbGxpbmcgb2YgcHJv Y2VzcyBpcyBjbG9zaW5nIGFsbCBvcGVuIGlub2RlcywgeW91IGNhbiBiZSBzdXJlIHRoYXQgZmls ZSBzeXN0ZW0gaXMgdW5tb3VudGVkIGlmIHRoZXJlIGFyZSBubyBvdGhlciBwcm9jZXNzZXMuIE9m IGNvdXJzZSB0aGlzIGlzIGEgcGxhY2UgZm9yIHJhY2UgY29uZGl0aW9ucy4gRm9yIGV4YW1wbGUs DQogdW5tb3VudGluZyBjYW4gYmUgc3R1Y2sgc29tZSB0aW1lIGlmIHVuZGVybHlpbmcgc3RvcmFn ZSBpcyB1bnJlc3BvbnNpdmUuIE11bHRpcGF0aCBzaG91bGQgZHJvcCBhbGwgSU9zIHdpdGggRUlP LiBWRFNNIGhhcyB0byB3YWl0IGVub3VnaCB0aW1lIGFmdGVyIGFjcXVpcmluZyB0aGUgbG9jayBi ZWZvcmUgbW91bnRpbmcgbWFzdGVyZnMuPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGJsb2Nr cXVvdGUgdHlwZT0iY2l0ZSI+DQo8ZGl2IGRpcj0ibHRyIj4NCjxkaXYgY2xhc3M9ImdtYWlsX3F1 b3RlIj4NCjxkaXY+RG8geW91IGtub3cgaWYgdGhlIGJlaGF2aW91ciBpcyBkb2N1bWVudGVkIHNv bWV3aGVyZT88L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Jsb2NrcXVvdGU+DQo8ZGl2Pjxicj4N CjwvZGl2Pg0KPGRpdj51bW91bnQoOCk6PC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2Pg0KPGRpdj4t bCwgLS1sYXp5PC9kaXY+DQo8ZGl2PkxhenkgdW5tb3VudC4mbmJzcDsmbmJzcDtEZXRhY2ggdGhl IGZpbGVzeXN0ZW0gZnJvbSB0aGUgZmlsZXN5c3RlbSBoaWVyYXJjaHkgbm93LCBhbmQgY2xlYW51 cCBhbGwgcmVmZXJlbmNlcyB0byB0aGUgZmlsZXN5c3RlbSBhcyBzb29uIGFzIGl0IGlzIG5vdCBi dXN5IGFueW1vcmUuJm5ic3A7Jm5ic3A7KFJlcXVpcmVzIGtlcm5lbCAyLjQuMTEgb3IgbGF0ZXIu KTwvZGl2Pg0KPGRpdj48L2Rpdj4NCjxkaXY+PGJyPg0KPC9kaXY+DQo8ZGl2Pjxicj4NCjwvZGl2 Pg0KPGRpdj4NCjxkaXYgZGlyPSJsdHIiPg0KPGRpdiBjbGFzcz0iZ21haWxfcXVvdGUiPjwvZGl2 Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYm9keT4NCjwvaHRtbD4NCg== --_000_14925401941105315camelacroniscom_--
Hi Nir, BMC - board management controller, in my case I have ilo. Yes I set up power management for all hosts - ovirt sees ilo status as ok. I use remote pdu to shutdown the port, after that happens the picture I attached. After I switch power port on, ovirt is able to read ilo status, sees that Linux is down and immediately switches the spm server. On Mon, Apr 17, 2017 at 6:07 AM Nir Soffer <nsoffer@redhat.com> wrote:
On Mon, Apr 17, 2017 at 8:24 AM Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
But actually, it didn't work well. After main SPM host went down I see this
[image: Screen Shot 2017-04-16 at 10.22.00 PM.png]
2017-04-17 05:23:15,554Z ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM Init: could not find reported vds or not up - pool: 'STG' vds_spm_id: '1' 2017-04-17 05:23:15,567Z INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] SPM selection - vds seems as spm 'tank5' 2017-04-17 05:23:15,567Z WARN [org.ovirt.engine.core.vdsbroker.irsbroker.IrsProxy] (DefaultQuartzScheduler5) [4dcc033d-26bf-49bb-bfaa-03a970dbbec1] spm vds is non responsive, stopping spm selection.
So that means only if BMC is up it's possible to automatically switch SPM host?
BMC?
If your SPM is no responsive, the system will try to fence it. Did you configure power management for all hosts? did you check that it work? How did you simulate non-responsive host?
If power management is not configured or fail, the system cannot move the spm to another host, unless you manually confirm that the SPM host was rebooted.
Nir
Thanks
On Sun, Apr 16, 2017 at 8:29 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Oh, fence agent works fine if I select ilo4, Thank you for your help!
On Sun, Apr 16, 2017 at 8:22 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:19 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Makes sense. I was trying to set it up, but doesn't work with our staging hardware. We have old ilo100, I'll try again. Thanks!
It is absolutely necessary for any HA to work properly. There's of course the "confirm host has been shutdown" option, which serves as an override for the fence command, but it's manual
On Sun, Apr 16, 2017 at 8:18 PM Dan Yasny <dyasny@gmail.com> wrote:
On Sun, Apr 16, 2017 at 11:15 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
> Fence agent under each node? >
When you configure a host, there's the power management tab, where you need to enter the bmc details for the host. If you don't have fencing enabled, how do you expect the system to make sure a host running a service is actually down (and it is safe to start HA services elsewhere), and not, for example, just unreachable by the engine? How do you avoid a splitbraid -> SBA ?
> > On Sun, Apr 16, 2017 at 8:14 PM Dan Yasny <dyasny@gmail.com> wrote: > >> On Sun, Apr 16, 2017 at 11:13 PM, Konstantin Raskoshnyi < >> konrasko@gmail.com> wrote: >> >>> "Corner cases"? >>> I tried to simulate crash of SPM server and ovirt kept trying to >>> reistablished connection to the failed node. >>> >> >> Did you configure fencing? >> >> >>> >>> >>> On Sun, Apr 16, 2017 at 8:10 PM Dan Yasny <dyasny@gmail.com> >>> wrote: >>> >>>> On Sun, Apr 16, 2017 at 7:29 AM, Nir Soffer <nsoffer@redhat.com> >>>> wrote: >>>> >>>>> On Sun, Apr 16, 2017 at 2:05 PM Dan Yasny <dyasny@redhat.com> >>>>> wrote: >>>>> >>>>>> >>>>>> >>>>>> On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> >>>>>> wrote: >>>>>> >>>>>> On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> >>>>>> wrote: >>>>>> >>>>>>> When you set up a storage domain, you need to specify a host >>>>>>> to perform the initial storage operations, but once the SD is defined, it's >>>>>>> details are in the engine database, and all the hosts get connected to it >>>>>>> directly. If the first host you used to define the SD goes down, all other >>>>>>> hosts will still remain connected and work. SPM is an HA service, and if >>>>>>> the current SPM host goes down, SPM gets started on another host in the DC. >>>>>>> In short, unless your actual NFS exporting host goes down, there is no >>>>>>> outage. >>>>>>> >>>>>> >>>>>> There is no storage outage, but if you shutdown the spm host, >>>>>> the spm host >>>>>> will not move to a new host until the spm host is online again, >>>>>> or you confirm >>>>>> manually that the spm host was rebooted. >>>>>> >>>>>> >>>>>> In a properly configured setup the SBA should take care of >>>>>> that. That's the whole point of HA services >>>>>> >>>>> >>>>> In some cases like power loss or hardware failure, there is no >>>>> way to start >>>>> the spm host, and the system cannot recover automatically. >>>>> >>>> >>>> There are always corner cases, no doubt. But in a normal >>>> situation. where an SPM host goes down because of a hardware failure, it >>>> gets fenced, other hosts contend for SPM and start it. No surprises there. >>>> >>>> >>>>> >>>>> Nir >>>>> >>>>> >>>>>> >>>>>> >>>>>> Nir >>>>>> >>>>>> >>>>>>> >>>>>>> On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < >>>>>>> konrasko@gmail.com> wrote: >>>>>>> >>>>>>>> Hi Fernando, >>>>>>>> I see each host has direct connection nfs mount, but yes, if >>>>>>>> main host to which I connected nfs storage going down the storage becomes >>>>>>>> unavailable and all vms are down >>>>>>>> >>>>>>>> >>>>>>>> On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < >>>>>>>> fernando.frediani@upx.com> wrote: >>>>>>>> >>>>>>>>> Hello Konstantin. >>>>>>>>> >>>>>>>>> That doesn`t make much sense make a whole cluster depend on >>>>>>>>> a single host. From what I know any host talk directly to NFS Storage Array >>>>>>>>> or whatever other Shared Storage you have. >>>>>>>>> Have you tested that host going down if that affects the >>>>>>>>> other with the NFS mounted directlly in a NFS Storage array ? >>>>>>>>> >>>>>>>>> Fernando >>>>>>>>> >>>>>>>>> 2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>>> konrasko@gmail.com>: >>>>>>>>> >>>>>>>>>> In ovirt you have to attach storage through specific host. >>>>>>>>>> If host goes down storage is not available. >>>>>>>>>> >>>>>>>>>> On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < >>>>>>>>>> fernando.frediani@upx.com> wrote: >>>>>>>>>> >>>>>>>>>>> Well, make it not go through host1 and dedicate a storage >>>>>>>>>>> server for running NFS and make both hosts connect to it. >>>>>>>>>>> In my view NFS is much easier to manage than any other >>>>>>>>>>> type of storage, specially FC and iSCSI and performance is pretty much the >>>>>>>>>>> same, so you won`t get better results other than management going to other >>>>>>>>>>> type. >>>>>>>>>>> >>>>>>>>>>> Fernando >>>>>>>>>>> >>>>>>>>>>> 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi < >>>>>>>>>>> konrasko@gmail.com>: >>>>>>>>>>> >>>>>>>>>>>> Hi guys, >>>>>>>>>>>> I have one nfs storage, >>>>>>>>>>>> it's connected through host1. >>>>>>>>>>>> host2 also has access to it, I can easily migrate >>>>>>>>>>>> vms between them. >>>>>>>>>>>> >>>>>>>>>>>> The question is - if host1 is down - all infrastructure >>>>>>>>>>>> is down, since all traffic goes through host1, >>>>>>>>>>>> is there any way in oVirt to use redundant storage? >>>>>>>>>>>> >>>>>>>>>>>> Only glusterfs? >>>>>>>>>>>> >>>>>>>>>>>> Thanks >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> _______________________________________________ >>>>>>>>>>>> Users mailing list >>>>>>>>>>>> Users@ovirt.org >>>>>>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Users mailing list >>>>>>>> Users@ovirt.org >>>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>>> >>>>>>>> >>>>>>> _______________________________________________ >>>>>>> Users mailing list >>>>>>> Users@ovirt.org >>>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Users mailing list >>>>>> Users@ovirt.org >>>>>> http://lists.ovirt.org/mailman/listinfo/users >>>>>> >>>>>> >>>>>>
What's SBA? On Sun, Apr 16, 2017 at 4:05 AM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
> Hi guys, > I have one nfs storage, > it's connected through host1. > host2 also has access to it, I can easily migrate vms between them. > > The question is - if host1 is down - all infrastructure is down, > since all traffic goes through host1, > is there any way in oVirt to use redundant storage? > > Only glusterfs? > > Thanks > > > _______________________________________________ > Users mailing list > Users@ovirt.org > http://lists.ovirt.org/mailman/listinfo/users > >
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Sun, Apr 16, 2017 at 12:44 PM, Konstantin Raskoshnyi <konrasko@gmail.com> wrote:
What's SBA?
split brain avoidance mechanism
On Sun, Apr 16, 2017 at 4:05 AM Dan Yasny <dyasny@redhat.com> wrote:
On Apr 16, 2017 7:01 AM, "Nir Soffer" <nsoffer@redhat.com> wrote:
On Sun, Apr 16, 2017 at 4:17 AM Dan Yasny <dyasny@gmail.com> wrote:
When you set up a storage domain, you need to specify a host to perform the initial storage operations, but once the SD is defined, it's details are in the engine database, and all the hosts get connected to it directly. If the first host you used to define the SD goes down, all other hosts will still remain connected and work. SPM is an HA service, and if the current SPM host goes down, SPM gets started on another host in the DC. In short, unless your actual NFS exporting host goes down, there is no outage.
There is no storage outage, but if you shutdown the spm host, the spm host will not move to a new host until the spm host is online again, or you confirm manually that the spm host was rebooted.
In a properly configured setup the SBA should take care of that. That's the whole point of HA services
Nir
On Sat, Apr 15, 2017 at 1:53 PM, Konstantin Raskoshnyi < konrasko@gmail.com> wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI < fernando.frediani@upx.com> wrote:
> Well, make it not go through host1 and dedicate a storage server for > running NFS and make both hosts connect to it. > In my view NFS is much easier to manage than any other type of > storage, specially FC and iSCSI and performance is pretty much the same, so > you won`t get better results other than management going to other type. > > Fernando > > 2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com> > : > >> Hi guys, >> I have one nfs storage, >> it's connected through host1. >> host2 also has access to it, I can easily migrate vms between them. >> >> The question is - if host1 is down - all infrastructure is down, >> since all traffic goes through host1, >> is there any way in oVirt to use redundant storage? >> >> Only glusterfs? >> >> Thanks >> >> >> _______________________________________________ >> Users mailing list >> Users@ovirt.org >> http://lists.ovirt.org/mailman/listinfo/users >> >>
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
This is a multi-part message in MIME format. --------------C778CD5E4FC1747C2A9FE3B2 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit If this works in this way then it's a huge downside in the architecture. Perhaps someone can clarify in more details. Fernando On 15/04/2017 14:53, Konstantin Raskoshnyi wrote:
Hi Fernando, I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down
On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Hello Konstantin.
That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have. Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?
Fernando
2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>>:
In ovirt you have to attach storage through specific host. If host goes down storage is not available.
On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <fernando.frediani@upx.com <mailto:fernando.frediani@upx.com>> wrote:
Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it. In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.
Fernando
2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <konrasko@gmail.com <mailto:konrasko@gmail.com>>:
Hi guys, I have one nfs storage, it's connected through host1. host2 also has access to it, I can easily migrate vms between them.
The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1, is there any way in oVirt to use redundant storage?
Only glusterfs?
Thanks
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
--------------C778CD5E4FC1747C2A9FE3B2 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit <html> <head> <meta content="text/html; charset=utf-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <p>If this works in this way then it's a huge downside in the architecture. Perhaps someone can clarify in more details.</p> <p>Fernando<br> </p> <br> <div class="moz-cite-prefix">On 15/04/2017 14:53, Konstantin Raskoshnyi wrote:<br> </div> <blockquote cite="mid:CAJU06NOAMePk_OaA1yAgeV0pHdb9qXR7Yh=Cui3Gixs=vtnT-Q@mail.gmail.com" type="cite"> <div>Hi Fernando,</div> <div>I see each host has direct connection nfs mount, but yes, if main host to which I connected nfs storage going down the storage becomes unavailable and all vms are down</div> <div><br> </div> <div><br> <div class="gmail_quote"> <div>On Sat, Apr 15, 2017 at 10:37 AM FERNANDO FREDIANI <<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com">fernando.frediani@upx.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div> <div>Hello Konstantin.<br> <br> </div> That doesn`t make much sense make a whole cluster depend on a single host. From what I know any host talk directly to NFS Storage Array or whatever other Shared Storage you have.<br> </div> Have you tested that host going down if that affects the other with the NFS mounted directlly in a NFS Storage array ?<br> <br> </div> </div> <div>Fernando<br> </div> <div class="gmail_extra"><br> <div class="gmail_quote">2017-04-15 12:42 GMT-03:00 Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span>:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>In ovirt you have to attach storage through specific host. </div> <div>If host goes down storage is not available. </div> <div><br> <div class="gmail_quote"> <div>On Sat, Apr 15, 2017 at 7:31 AM FERNANDO FREDIANI <<a moz-do-not-send="true" href="mailto:fernando.frediani@upx.com" target="_blank">fernando.frediani@upx.com</a>> wrote:<br> </div> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div> <div> <div>Well, make it not go through host1 and dedicate a storage server for running NFS and make both hosts connect to it.<br> </div> In my view NFS is much easier to manage than any other type of storage, specially FC and iSCSI and performance is pretty much the same, so you won`t get better results other than management going to other type.<br> <br> </div> Fernando<br> </div> <div class="gmail_extra"><br> </div> <div class="gmail_extra"> <div class="gmail_quote">2017-04-15 5:25 GMT-03:00 Konstantin Raskoshnyi <span><<a moz-do-not-send="true" href="mailto:konrasko@gmail.com" target="_blank">konrasko@gmail.com</a>></span>:<br> </div> </div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <div>Hi guys, <div>I have one nfs storage,</div> <div>it's connected through host1.</div> <div>host2 also has access to it, I can easily migrate vms between them.</div> <div><br> </div> <div>The question is - if host1 is down - all infrastructure is down, since all traffic goes through host1,</div> <div>is there any way in oVirt to use redundant storage?</div> <div><br> </div> <div>Only glusterfs?</div> <div><br> </div> <div>Thanks</div> <div><br> </div> </div> <br> </blockquote> </div> </div> <div class="gmail_extra"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">_______________________________________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" target="_blank">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" rel="noreferrer" target="_blank">http://lists.ovirt.org/mailman/listinfo/users</a><br> <br> </blockquote> </div> </div> </blockquote> </div> </div> </blockquote> </div> <br> </div> </blockquote> </div> </div> </blockquote> <br> </body> </html> --------------C778CD5E4FC1747C2A9FE3B2--
participants (9)
-
Adam Litke -
Charles Kozler -
Dan Yasny -
Dan Yasny -
FERNANDO FREDIANI -
Konstantin Raskoshnyi -
Nir Soffer -
Niyazi Elvan -
Pavel Gashev