Involuntary emergency upgrade from 4.1 to 4.2
I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup. The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups. The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something. Any ideas?
Il giorno lun 29 ott 2018 alle ore 08:49 <djohnson@maxistechnology.com> ha scritto:
I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas?
Douglas, Didi, can you please assist here?
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
-- SANDRO BONAZZOLA MANAGER, SOFTWARE ENGINEERING, EMEA R&D RHV Red Hat EMEA <https://www.redhat.com/> sbonazzo@redhat.com <https://red.ht/sig> <https://www.redhat.com/en/events/red-hat-open-source-day-italia?sc_cid=701f2000000RgRyAAK>
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe you could find some issues with other repos. If you want to take a shortcut, https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance. You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms. On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
Thanks, I'll check it out. On Mon, Oct 29, 2018, 2:54 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe you could find some issues with other repos.
If you want to take a shortcut, https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance. You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe you could find some issues with other repos.
If you want to take a shortcut, https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance. You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real. See also e.g.: https://lists.ovirt.org/pipermail/users/2017-March/080346.html https://bugzilla.redhat.com/show_bug.cgi?id=1425788 That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently. Good luck and best regards,
The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
Thank you for your generous help. 4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset. I will give these suggestions a try On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe
you could find some issues with other repos.
If you want to take a shortcut,
https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance.
You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware
failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old, but
was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
The storage domains are all intact (I think), and the hosts are still
running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
I understand your sense of panic, I accidentally destroyed my RHEV manager a few months ago, and it was quite a bad time. It sounds like you have some good suggestions on how to recover, I just wanted to share some experience with you. Your hypervisors will happily continue running your vm's, as long as nothing interrupts them. If a guest shuts down or hangs, you'll have no way to start it back up or reboot it. likewise, if a hypervisor crashes, you wont be able to start it's guests back up. Your storage domains should be importable into 4.2 from 4.1. One of the clusters I'd broken was fixed that way. I did have to log into each guest and shut it down before I could safely bring the storage domain into my new manager though. I was going from a broken 3.6 manager to a new 4.1 manager in that case. Not sure if its different from 4.1 to 4.2. The procedure went something like this, from my memory: Build a new 4.1 manager. Shut down vm's on your rhev/oVirt hypervisors. Reinstall one of the hypervisors, and join it to the new manager. Attach your storage domains using the Import Domain button, not the New domain button. Domain should come in, and have all of your vms listed. Reinstall the rest of the hypervisors, and join them to the new manager. I believe I had to activate, or import each VM as well. It's been several months since I did all that. I'm not sure if you're working with upstream oVirt, or RHV, i'm on RHV. My 4.1 repos are certainly still available from Red Hat. Not sure about the status of oVirt repos. On Mon, Oct 29, 2018 at 10:27 AM David Johnson <djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe
you could find some issues with other repos.
If you want to take a shortcut,
https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance.
You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware
failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old,
but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
The storage domains are all intact (I think), and the hosts are still
running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/J2BB32HKODBY4C...
Thank you everyone for all of your help. Here is where things stand now: I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru. I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example. As near as I can tell, the next crucial step is to import the existing storage domains. There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information. Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center? Thank you in advance, On Mon, Oct 29, 2018 at 7:14 AM David Johnson <djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe
you could find some issues with other repos.
If you want to take a shortcut,
https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance.
You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware
failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old,
but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
The storage domains are all intact (I think), and the hosts are still
running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
The cluster will need a default storage domain before you can import your existing domain. That much I do remember. What's causing your error I don't know. What sort of storage are you working with? On Mon, Oct 29, 2018, 6:04 PM David Johnson <djohnson@maxistechnology.com> wrote:
Thank you everyone for all of your help. Here is where things stand now:
I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru.
I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example.
As near as I can tell, the next crucial step is to import the existing storage domains.
There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information.
Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center?
Thank you in advance,
On Mon, Oct 29, 2018 at 7:14 AM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there,
maybe you could find some issues with other repos.
If you want to take a shortcut,
https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance.
You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a hardware
failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old,
but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
The storage domains are all intact (I think), and the hosts are still
running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CN4Y3CC6XKROOF...
Hi everyone, and thank you again for all of your help. Here is the latest update in the never ending story. I worked through NFS woes on my san, configured a throwaway storage domain to bootstrap the system, and have reconnected the storage domains. The main data domain is in "Maintenance" state Host 1 is in "Preparing for Maintenance" state with SPM = "normal", Host 2 is in "Up" state with SPM = "SPM", I think the next step is to get host 1 either fully into or out of maintenance mode, and the storage domain out of maintenance mode. Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com [image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com *stay connected <http://www.linkedin.com/in/pojoguy>* On Mon, Oct 29, 2018 at 10:10 PM Nathan Lager <lagern@lafayette.edu> wrote:
The cluster will need a default storage domain before you can import your existing domain. That much I do remember. What's causing your error I don't know. What sort of storage are you working with?
On Mon, Oct 29, 2018, 6:04 PM David Johnson <djohnson@maxistechnology.com> wrote:
Thank you everyone for all of your help. Here is where things stand now:
I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru.
I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example.
As near as I can tell, the next crucial step is to import the existing storage domains.
There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information.
Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center?
Thank you in advance,
On Mon, Oct 29, 2018 at 7:14 AM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi <stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there,
maybe you could find some issues with other repos.
If you want to take a shortcut,
https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance.
You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote:
I'm looking for a hand in recovering my ovirt cluster from a
hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup.
The problem is, the cluster was 4.1, which is less than a year old,
but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
The storage domains are all intact (I think), and the hosts are
still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something.
Any ideas? _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CN4Y3CC6XKROOF...
Host 1 is now back in "normal" mode. It was stuck in preparing for maintenance because the VM's were not fully under the control of the cluster. Once I powered off the VM's it finished what it was doing, rebooted, and is now properly part of the cluster. However, the main data domain is still in "Maintenance" state, and I can't find any way to make it "Active". My ISO partition is also inactive, with no visible way to make it "Active". Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com [image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com *stay connected <http://www.linkedin.com/in/pojoguy>* On Tue, Oct 30, 2018 at 4:22 PM David Johnson <djohnson@maxistechnology.com> wrote:
Hi everyone, and thank you again for all of your help.
Here is the latest update in the never ending story.
I worked through NFS woes on my san, configured a throwaway storage domain to bootstrap the system, and have reconnected the storage domains.
The main data domain is in "Maintenance" state Host 1 is in "Preparing for Maintenance" state with SPM = "normal", Host 2 is in "Up" state with SPM = "SPM",
I think the next step is to get host 1 either fully into or out of maintenance mode, and the storage domain out of maintenance mode.
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Mon, Oct 29, 2018 at 10:10 PM Nathan Lager <lagern@lafayette.edu> wrote:
The cluster will need a default storage domain before you can import your existing domain. That much I do remember. What's causing your error I don't know. What sort of storage are you working with?
On Mon, Oct 29, 2018, 6:04 PM David Johnson <djohnson@maxistechnology.com> wrote:
Thank you everyone for all of your help. Here is where things stand now:
I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru.
I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example.
As near as I can tell, the next crucial step is to import the existing storage domains.
There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information.
Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center?
Thank you in advance,
On Mon, Oct 29, 2018 at 7:14 AM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi < stirabos@redhat.com> wrote:
Hi, AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there,
maybe you could find some issues with other repos.
If you want to take a shortcut,
https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance.
You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms.
On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote: > > I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup. > > The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
> > The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something. > > Any ideas? > _______________________________________________ > Users mailing list -- users@ovirt.org > To unsubscribe send an email to users-leave@ovirt.org > Privacy Statement: https://www.ovirt.org/site/privacy-policy/ > oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ > List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CN4Y3CC6XKROOF...
The saga has ended and all is well. When I stopped all of the active VM's, the storage domains came back to normal and came online. At that point I had to delete the VM's that had been running unmanaged on the hosts, then I could import all of the objects from he storage domain into the data center. the only thing I lost was a bitnami redmine VM that had been installed from an OVA. This was no major loss, since I backed it up at the beginning of this adventure. In retrospect, the biggest issues I had were: * being overly cautious because I didn't know which operations were destructive, and * finding the right actions in the GUI when I decided to take the plunge. Thank you all for your generous assistance. Regards, David Johnson On Tue, Oct 30, 2018 at 10:29 PM David Johnson <djohnson@maxistechnology.com> wrote:
Host 1 is now back in "normal" mode. It was stuck in preparing for maintenance because the VM's were not fully under the control of the cluster. Once I powered off the VM's it finished what it was doing, rebooted, and is now properly part of the cluster.
However, the main data domain is still in "Maintenance" state, and I can't find any way to make it "Active". My ISO partition is also inactive, with no visible way to make it "Active".
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Tue, Oct 30, 2018 at 4:22 PM David Johnson < djohnson@maxistechnology.com> wrote:
Hi everyone, and thank you again for all of your help.
Here is the latest update in the never ending story.
I worked through NFS woes on my san, configured a throwaway storage domain to bootstrap the system, and have reconnected the storage domains.
The main data domain is in "Maintenance" state Host 1 is in "Preparing for Maintenance" state with SPM = "normal", Host 2 is in "Up" state with SPM = "SPM",
I think the next step is to get host 1 either fully into or out of maintenance mode, and the storage domain out of maintenance mode.
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Mon, Oct 29, 2018 at 10:10 PM Nathan Lager <lagern@lafayette.edu> wrote:
The cluster will need a default storage domain before you can import your existing domain. That much I do remember. What's causing your error I don't know. What sort of storage are you working with?
On Mon, Oct 29, 2018, 6:04 PM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you everyone for all of your help. Here is where things stand now:
I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru.
I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example.
As near as I can tell, the next crucial step is to import the existing storage domains.
There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information.
Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center?
Thank you in advance,
On Mon, Oct 29, 2018 at 7:14 AM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi < stirabos@redhat.com> wrote: > > Hi, > AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, maybe you could find some issues with other repos. > > If you want to take a shortcut, https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... contains latest 4.1 based engine appliance. > You can extract the OVA file from there and use it to boot a temporary VM with virt-install and then upgrade the engine there to 4.2 in order to take a new 4.2 based backup that your can restore where you prefer with up to date rpms. > > > On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> wrote: >> >> I'm looking for a hand in recovering my ovirt cluster from a hardware failure. The hard drive on my cluster controller failed, and I would like to recover from backup. >> >> The problem is, the cluster was 4.1, which is less than a year old, but was nevertheless removed from the active repositories back in May. 4.2 will not recover from 4.1 backups.
If you are brave, you can also "cheat" - patch engine-backup to allow recovering 4.1. That's a trivial patch, and the main problem with it is that no-one tested it, and I do expect it might introduce subtle issues. But considering the alternatives, it might be a reasonable approach. If you do, try restoring first on an _isolated_ VM somewhere, try to see how the engine behaves after restore (it will not work very well, because it will not manage to access its hosts - if you indeed isolated it well enough), and if it looks ok, try for real.
See also e.g.:
https://lists.ovirt.org/pipermail/users/2017-March/080346.html
https://bugzilla.redhat.com/show_bug.cgi?id=1425788
That said, not sure why 4.1 does not work for you. I think it should still work, although I didn't try by myself recently.
Good luck and best regards,
>> >> The storage domains are all intact (I think), and the hosts are still running (unmanaged). I've tried to manually restore the engine from backups, but either the upgrade is reinitializing or I am missing something. >> >> Any ideas? >> _______________________________________________ >> Users mailing list -- users@ovirt.org >> To unsubscribe send an email to users-leave@ovirt.org >> Privacy Statement: https://www.ovirt.org/site/privacy-policy/ >> oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ >> List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M... > > _______________________________________________ > Users mailing list -- users@ovirt.org > To unsubscribe send an email to users-leave@ovirt.org > Privacy Statement: https://www.ovirt.org/site/privacy-policy/ > oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ > List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52...
-- Didi
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CN4Y3CC6XKROOF...
On Thu, Nov 1, 2018 at 6:13 AM David Johnson <djohnson@maxistechnology.com> wrote:
The saga has ended and all is well.
When I stopped all of the active VM's, the storage domains came back to normal and came online. At that point I had to delete the VM's that had been running unmanaged on the hosts, then I could import all of the objects from he storage domain into the data center. the only thing I lost was a bitnami redmine VM that had been installed from an OVA. This was no major loss, since I backed it up at the beginning of this adventure.
In retrospect, the biggest issues I had were: * being overly cautious because I didn't know which operations were destructive, and * finding the right actions in the GUI when I decided to take the plunge.
Thank you all for your generous assistance.
Glad to hear that, thanks for the report! Best regards,
Regards, David Johnson
On Tue, Oct 30, 2018 at 10:29 PM David Johnson < djohnson@maxistechnology.com> wrote:
Host 1 is now back in "normal" mode. It was stuck in preparing for maintenance because the VM's were not fully under the control of the cluster. Once I powered off the VM's it finished what it was doing, rebooted, and is now properly part of the cluster.
However, the main data domain is still in "Maintenance" state, and I can't find any way to make it "Active". My ISO partition is also inactive, with no visible way to make it "Active".
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Tue, Oct 30, 2018 at 4:22 PM David Johnson < djohnson@maxistechnology.com> wrote:
Hi everyone, and thank you again for all of your help.
Here is the latest update in the never ending story.
I worked through NFS woes on my san, configured a throwaway storage domain to bootstrap the system, and have reconnected the storage domains.
The main data domain is in "Maintenance" state Host 1 is in "Preparing for Maintenance" state with SPM = "normal", Host 2 is in "Up" state with SPM = "SPM",
I think the next step is to get host 1 either fully into or out of maintenance mode, and the storage domain out of maintenance mode.
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Mon, Oct 29, 2018 at 10:10 PM Nathan Lager <lagern@lafayette.edu> wrote:
The cluster will need a default storage domain before you can import your existing domain. That much I do remember. What's causing your error I don't know. What sort of storage are you working with?
On Mon, Oct 29, 2018, 6:04 PM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you everyone for all of your help. Here is where things stand now:
I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru.
I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example.
As near as I can tell, the next crucial step is to import the existing storage domains.
There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information.
Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center?
Thank you in advance,
On Mon, Oct 29, 2018 at 7:14 AM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
> On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi < > stirabos@redhat.com> wrote: > > > > Hi, > > AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, > maybe you could find some issues with other repos. > > > > If you want to take a shortcut, > https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... > contains latest 4.1 based engine appliance. > > You can extract the OVA file from there and use it to boot a > temporary VM with virt-install and then upgrade the engine there to 4.2 in > order to take a new 4.2 based backup that your can restore where you prefer > with up to date rpms. > > > > > > On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> > wrote: > >> > >> I'm looking for a hand in recovering my ovirt cluster from a > hardware failure. The hard drive on my cluster controller failed, and I > would like to recover from backup. > >> > >> The problem is, the cluster was 4.1, which is less than a year > old, but was nevertheless removed from the active repositories back in May. > 4.2 will not recover from 4.1 backups. > > If you are brave, you can also "cheat" - patch engine-backup to allow > recovering 4.1. That's a trivial patch, and the main problem with it > is that no-one tested it, and I do expect it might introduce subtle > issues. But considering the alternatives, it might be a reasonable > approach. If you do, try restoring first on an _isolated_ VM > somewhere, > try to see how the engine behaves after restore (it will not work > very well, because it will not manage to access its hosts - if you > indeed isolated it well enough), and if it looks ok, try for real. > > See also e.g.: > > https://lists.ovirt.org/pipermail/users/2017-March/080346.html > > https://bugzilla.redhat.com/show_bug.cgi?id=1425788 > > That said, not sure why 4.1 does not work for you. I think it should > still work, although I didn't try by myself recently. > > Good luck and best regards, > > >> > >> The storage domains are all intact (I think), and the hosts are > still running (unmanaged). I've tried to manually restore the engine from > backups, but either the upgrade is reinitializing or I am missing something. > >> > >> Any ideas? > >> _______________________________________________ > >> Users mailing list -- users@ovirt.org > >> To unsubscribe send an email to users-leave@ovirt.org > >> Privacy Statement: https://www.ovirt.org/site/privacy-policy/ > >> oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > >> List Archives: > https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M... > > > > _______________________________________________ > > Users mailing list -- users@ovirt.org > > To unsubscribe send an email to users-leave@ovirt.org > > Privacy Statement: https://www.ovirt.org/site/privacy-policy/ > > oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > > List Archives: > https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52... > > > > -- > Didi > _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CN4Y3CC6XKROOF...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/76YBSLTDI73EMI...
-- Didi
So glad to hear it has ended well! On Thu, Nov 1, 2018 at 12:21 AM David Johnson <djohnson@maxistechnology.com> wrote:
The saga has ended and all is well.
When I stopped all of the active VM's, the storage domains came back to normal and came online. At that point I had to delete the VM's that had been running unmanaged on the hosts, then I could import all of the objects from he storage domain into the data center. the only thing I lost was a bitnami redmine VM that had been installed from an OVA. This was no major loss, since I backed it up at the beginning of this adventure.
In retrospect, the biggest issues I had were: * being overly cautious because I didn't know which operations were destructive, and * finding the right actions in the GUI when I decided to take the plunge.
Thank you all for your generous assistance.
Regards, David Johnson
On Tue, Oct 30, 2018 at 10:29 PM David Johnson < djohnson@maxistechnology.com> wrote:
Host 1 is now back in "normal" mode. It was stuck in preparing for maintenance because the VM's were not fully under the control of the cluster. Once I powered off the VM's it finished what it was doing, rebooted, and is now properly part of the cluster.
However, the main data domain is still in "Maintenance" state, and I can't find any way to make it "Active". My ISO partition is also inactive, with no visible way to make it "Active".
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Tue, Oct 30, 2018 at 4:22 PM David Johnson < djohnson@maxistechnology.com> wrote:
Hi everyone, and thank you again for all of your help.
Here is the latest update in the never ending story.
I worked through NFS woes on my san, configured a throwaway storage domain to bootstrap the system, and have reconnected the storage domains.
The main data domain is in "Maintenance" state Host 1 is in "Preparing for Maintenance" state with SPM = "normal", Host 2 is in "Up" state with SPM = "SPM",
I think the next step is to get host 1 either fully into or out of maintenance mode, and the storage domain out of maintenance mode.
Regards, David Johnson Director of Development, Maxis Technology 844.696.2947 ext 702 (o) | 479.531.3590 (c) djohnson@maxistechnology.com
[image: Maxis Techncology] <http://www.maxistechnology.com> www.maxistechnology.com
*stay connected <http://www.linkedin.com/in/pojoguy>*
On Mon, Oct 29, 2018 at 10:10 PM Nathan Lager <lagern@lafayette.edu> wrote:
The cluster will need a default storage domain before you can import your existing domain. That much I do remember. What's causing your error I don't know. What sort of storage are you working with?
On Mon, Oct 29, 2018, 6:04 PM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you everyone for all of your help. Here is where things stand now:
I gave up trying to recover from backup. I wasn't able to mount the OVA. Since this is a test rack, I looked at the amount of time I've sunk into it already (about the same as it took to build from scratch the first time around) and decided that it was worthwhile to risk starting from scratch rather than become an overnight bit bashing Ovirt guru.
I installed the 4.2 controller, upgraded the hosts to 4.2, and then added the two hosts to the default data center. At this point, all of the running JVM's were visible to the controller, but the vm's do not appear to be managed by the controller. I can't migrate them from one host to the other, for example.
As near as I can tell, the next crucial step is to import the existing storage domains.
There is no visible way to import an existing storage domain to an uninitialized data center (uninitialized data centers are not available to the Import Domain button)), so I created a new share on the SAN for a bootstrap domain, and am unable to connect to it. The error message given by OVirt is "Error while executing action Add Storage Connection: Problem while trying to mount target", which is not really information.
Is there a command line tool for importing the existing storage domains that will not choke on an uninitialized data center?
Thank you in advance,
On Mon, Oct 29, 2018 at 7:14 AM David Johnson < djohnson@maxistechnology.com> wrote:
Thank you for your generous help.
4.1 wouldn't install because a number of dependencies were pointing to dead links. It felt like over half, although I'm sure it was a more limited subset.
I will give these suggestions a try
On Mon, Oct 29, 2018, 4:50 AM Yedidyah Bar David <didi@redhat.com> wrote:
> On Mon, Oct 29, 2018 at 10:00 AM Simone Tiraboschi < > stirabos@redhat.com> wrote: > > > > Hi, > > AFAIK https://resources.ovirt.org/pub/ovirt-4.1/ is still there, > maybe you could find some issues with other repos. > > > > If you want to take a shortcut, > https://resources.ovirt.org/pub/ovirt-4.1/rpm/el7/noarch/ovirt-engine-applia... > contains latest 4.1 based engine appliance. > > You can extract the OVA file from there and use it to boot a > temporary VM with virt-install and then upgrade the engine there to 4.2 in > order to take a new 4.2 based backup that your can restore where you prefer > with up to date rpms. > > > > > > On Mon, Oct 29, 2018 at 8:49 AM <djohnson@maxistechnology.com> > wrote: > >> > >> I'm looking for a hand in recovering my ovirt cluster from a > hardware failure. The hard drive on my cluster controller failed, and I > would like to recover from backup. > >> > >> The problem is, the cluster was 4.1, which is less than a year > old, but was nevertheless removed from the active repositories back in May. > 4.2 will not recover from 4.1 backups. > > If you are brave, you can also "cheat" - patch engine-backup to allow > recovering 4.1. That's a trivial patch, and the main problem with it > is that no-one tested it, and I do expect it might introduce subtle > issues. But considering the alternatives, it might be a reasonable > approach. If you do, try restoring first on an _isolated_ VM > somewhere, > try to see how the engine behaves after restore (it will not work > very well, because it will not manage to access its hosts - if you > indeed isolated it well enough), and if it looks ok, try for real. > > See also e.g.: > > https://lists.ovirt.org/pipermail/users/2017-March/080346.html > > https://bugzilla.redhat.com/show_bug.cgi?id=1425788 > > That said, not sure why 4.1 does not work for you. I think it should > still work, although I didn't try by myself recently. > > Good luck and best regards, > > >> > >> The storage domains are all intact (I think), and the hosts are > still running (unmanaged). I've tried to manually restore the engine from > backups, but either the upgrade is reinitializing or I am missing something. > >> > >> Any ideas? > >> _______________________________________________ > >> Users mailing list -- users@ovirt.org > >> To unsubscribe send an email to users-leave@ovirt.org > >> Privacy Statement: https://www.ovirt.org/site/privacy-policy/ > >> oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > >> List Archives: > https://lists.ovirt.org/archives/list/users@ovirt.org/message/HRIUSVMGGCIG5M... > > > > _______________________________________________ > > Users mailing list -- users@ovirt.org > > To unsubscribe send an email to users-leave@ovirt.org > > Privacy Statement: https://www.ovirt.org/site/privacy-policy/ > > oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > > List Archives: > https://lists.ovirt.org/archives/list/users@ovirt.org/message/3OIB2WQSMHVM52... > > > > -- > Didi > _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/CN4Y3CC6XKROOF...
_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/76YBSLTDI73EMI...
participants (6)
-
David Johnson -
djohnson＠maxistechnology.com
-
Nathan Lager -
Sandro Bonazzola -
Simone Tiraboschi -
Yedidyah Bar David