- Users - oVirt List Archives

FOSDEM 2018 - CFP is about to close!
by Doron Fediuck 29 Nov '17

29 Nov '17

Hi all, FOSDEM 18's Virt and IaaS CFP ends on Friday! This is your last chance to submit your session, so do not wait any longer. You can find more details in [1]. Thanks, Doron [1] https://www.ovirt.org/blog/2017/10/come-to-fosdem-event/

1 0

strange problems with engine
by alex＠rt14.ru 29 Nov '17

29 Nov '17

2 2

Update gluster HCI from 4.1.3 to 4.1.7
by Gianluca Cecchi 28 Nov '17

28 Nov '17

Hello, I have an environment with 3 hosts and gluster HCI on 4.1.3. I'm following this link to take it to 4.1.7 https://www.ovirt.org/documentation/how-to/hosted-engine/#upgrade-hosted-en… The hosts and engine were at 7.3 prior of beginning the update. All went ok for the engine that now is on 7.4 (not rebooted yet) Points 4. 5. 6. for the first updated host were substituted by rebooting it I'm at point: 7. Exit the global maintenance mode: in a few minutes the engine VM should migrate to the fresh upgraded host cause it will get an higher score One note: actually exiting from global maintenance doesn't imply that the host previously put into maintenance exiting from it, correct? So in my workflow, before point 7., actually I have selected the host and activated it. Currently the situation is this - engine running on ovirt02 - update happened on ovirt03 Then after exiting from global maintenance I don't see the engine vm migrating to it. And in fact (see below) the score of ovirt02 is the same (3400) as the one of ovirt03, so it seems it is correct that engine remains there...? Which kind of messages should I see inside logs of engine/hosts? [root@ovirt01 ~]# rpm -q vdsm vdsm-4.19.20-1.el7.centos.x86_64 [root@ovirt02 ~]# rpm -q vdsm vdsm-4.19.20-1.el7.centos.x86_64 [root@ovirt02 ~]# [root@ovirt03 ~]# rpm -q vdsm vdsm-4.19.37-1.el7.centos.x86_64 from host ovirt01: [root@ovirt01 ~]# hosted-engine --vm-status --== Host 1 status ==-- conf_on_shared_storage : True Status up-to-date : True Hostname : ovirt01.localdomain.local Host ID : 1 Engine status : {"reason": "vm not running on this host", "health": "bad", "vm": "down", "detail": "unknown"} Score : 3352 stopped : False Local maintenance : False crc32 : 256f2128 local_conf_timestamp : 12251210 Host timestamp : 12251178 Extra metadata (valid at timestamp): metadata_parse_version=1 metadata_feature_version=1 timestamp=12251178 (Tue Nov 28 10:11:20 2017) host-id=1 score=3352 vm_conf_refresh_time=12251210 (Tue Nov 28 10:11:52 2017) conf_on_shared_storage=True maintenance=False state=EngineDown stopped=False --== Host 2 status ==-- conf_on_shared_storage : True Status up-to-date : True Hostname : 192.168.150.103 Host ID : 2 Engine status : {"health": "good", "vm": "up", "detail": "up"} Score : 3400 stopped : False Local maintenance : False crc32 : 9b8c8a6c local_conf_timestamp : 12219386 Host timestamp : 12219357 Extra metadata (valid at timestamp): metadata_parse_version=1 metadata_feature_version=1 timestamp=12219357 (Tue Nov 28 10:11:23 2017) host-id=2 score=3400 vm_conf_refresh_time=12219386 (Tue Nov 28 10:11:52 2017) conf_on_shared_storage=True maintenance=False state=EngineUp stopped=False --== Host 3 status ==-- conf_on_shared_storage : True Status up-to-date : True Hostname : ovirt03.localdomain.local Host ID : 3 Engine status : {"reason": "vm not running on this host", "health": "bad", "vm": "down", "detail": "unknown"} Score : 3400 stopped : False Local maintenance : False crc32 : 9f6399ef local_conf_timestamp : 2136 Host timestamp : 2136 Extra metadata (valid at timestamp): metadata_parse_version=1 metadata_feature_version=1 timestamp=2136 (Tue Nov 28 10:11:56 2017) host-id=3 score=3400 vm_conf_refresh_time=2136 (Tue Nov 28 10:11:56 2017) conf_on_shared_storage=True maintenance=False state=EngineDown stopped=False [root@ovirt01 ~]# Can I manually migrate engine vm to ovirt03? On ovirt03: [root@ovirt03 ~]# gluster volume info engine Volume Name: engine Type: Replicate Volume ID: 6e2bd1d7-9c8e-4c54-9d85-f36e1b871771 Status: Started Snapshot Count: 0 Number of Bricks: 1 x (2 + 1) = 3 Transport-type: tcp Bricks: Brick1: ovirt01.localdomain.local:/gluster/brick1/engine Brick2: ovirt02.localdomain.local:/gluster/brick1/engine Brick3: ovirt03.localdomain.local:/gluster/brick1/engine (arbiter) Options Reconfigured: performance.strict-o-direct: on nfs.disable: on user.cifs: off network.ping-timeout: 30 cluster.shd-max-threads: 6 cluster.shd-wait-qlength: 10000 cluster.locking-scheme: granular cluster.data-self-heal-algorithm: full performance.low-prio-threads: 32 features.shard-block-size: 512MB features.shard: on storage.owner-gid: 36 storage.owner-uid: 36 cluster.server-quorum-type: server cluster.quorum-type: auto network.remote-dio: off cluster.eager-lock: enable performance.stat-prefetch: off performance.io-cache: off performance.read-ahead: off performance.quick-read: off performance.readdir-ahead: on transport.address-family: inet [root@ovirt03 ~]# [root@ovirt03 ~]# gluster volume heal engine info Brick ovirt01.localdomain.local:/gluster/brick1/engine Status: Connected Number of entries: 0 Brick ovirt02.localdomain.local:/gluster/brick1/engine Status: Connected Number of entries: 0 Brick ovirt03.localdomain.local:/gluster/brick1/engine Status: Connected Number of entries: 0 [root@ovirt03 ~]# Thanks, Gianluca

2 2

"Cannot acquire bridge address" / Failure in "hosted-engine --deploy" on 4.1.8
by Frank Wall 28 Nov '17

28 Nov '17

Hi, when running "hosted-engine --deploy" on a fresh CentOS 7.4 with pre-built network configuration (and ovirtmgmt bridge), the following error occurs: 2017-11-27 19:08:12 DEBUG otopi.context context._executeMethod:142 method exception Traceback (most recent call last): File "/usr/lib/python2.7/site-packages/otopi/context.py", line 132, in _executeMethod method['method']() File "/usr/share/ovirt-hosted-engine-setup/scripts/../plugins/gr-he-common/network/bridge.py", line 295, in _get_hostname_from_bridge_if raise RuntimeError(_('Cannot acquire bridge address')) RuntimeError: Cannot acquire bridge address While checking the source of bridge.py line 295 the root cause is interesting: Running "vdsClient -s $HOSTNAME getVdsCapabilities" works perfectly, all NICs and bridges are returned, but only "networks" is returned as empty list: {}. Clearly looks like a bug. How can I resolve this issue? ovirt-hosted-engine-setup-2.1.4-1.el7.centos.noarch vdsm-4.19.40-1.el7.centos.x86_64 Regards - Frank

2 3

Time management in self hosted engine vm
by Gianluca Cecchi 28 Nov '17

28 Nov '17

Hello, I have a self hosted engine HCI environment that was born in 4.0.5 through the ansible/gdeploy mechanism and then gradually updated to 4.1.7 Now the engine and host are at CentOS 7.4 level. Inside the gdeploy jobs the engine vm was installed through the appliance at the 4.0.5 time. I see in web admin gui that hw clock of engine vm is configured as UTC (so far so good) and the same is inside the OS. And I would like to change it to UTC+1 like the 3 hosts I also see that neither chronyd nor ntpd are installed on engine vm. Can I fix the situation the normal os level way: - timedatectl set-timezone "Europe/Rome" - install and configure chronyd or better not to configure time sync in engine vm? Thanks, Gianluca

2 1

HA on Non Responsive
by Arthur Melo 28 Nov '17

28 Nov '17

Hello, there! Is it possible to reduce time to auto migrate machines when node is non responsive? Atenciosamente, Arthur Melo Linux User #302250

3 5

Hosts been evacuated unnecessarily
by FERNANDO FREDIANI 28 Nov '17

28 Nov '17

This is a multi-part message in MIME format. --------------85B0C37C08377D8D314D9F67 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Hello folks. Ou oVirt (4.1.7.3-1.el7.centos) which runs in one Datacenter and controls Nodes locally and also remotelly lost communication with the remote Nodes in another Datacenter. To this point nothing wrong as the Nodes can continue working as expected and running their Virtual Machines each without dependency of the oVirt Engine. What happened at some point is that when the communication between Engine and Hosts came back Hosts in the remote Datacenter got confused and initiated a Live Migration of ALL VMs from one of the hosts to another. I had also to restart vdsmd agent on all Hosts in order to get sanity my environment. What adds up even more strangeness to this scenario is that one of the Hosts affected by the need of restarting VDSM doesn't belong to the same Cluster as the others and had to have the vdsmd restarted. I understand the Hosts can survive without the Engine online with reduced possibilities but can communicated between them, but without affecting the VMs or even needing to do what happened in this scenario. Am I wrong on any of the assumptions ? Fernando --------------85B0C37C08377D8D314D9F67 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 7bit <html> <head> <meta http-equiv="content-type" content="text/html; charset=utf-8"> </head> <body bgcolor="#FFFFFF" text="#000000"> Hello folks. Ou oVirt (4.1.7.3-1.el7.centos) which runs in one Datacenter and controls Nodes locally and also remotelly lost communication with the remote Nodes in another Datacenter. To this point nothing wrong as the Nodes can continue working as expected and running their Virtual Machines each without dependency of the oVirt Engine. What happened at some point is that when the communication between Engine and Hosts came back Hosts in the remote Datacenter got confused and initiated a Live Migration of ALL VMs from one of the hosts to another. I had also to restart vdsmd agent on all Hosts in order to get sanity my environment. What adds up even more strangeness to this scenario is that one of the Hosts affected by the need of restarting VDSM doesn't belong to the same Cluster as the others and had to have the vdsmd restarted. I understand the Hosts can survive without the Engine online with reduced possibilities but can communicated between them, but without affecting the VMs or even needing to do what happened in this scenario. Am I wrong on any of the assumptions ? Fernando </body> </html> --------------85B0C37C08377D8D314D9F67--

1 0

OVS DPDK Performance
by FERNANDO FREDIANI 28 Nov '17

28 Nov '17

This is a multi-part message in MIME format. --------------BDE85DB92A8FD76EF257B82B Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Hello As some may have seen recently OVS DPDK has been introduced to oVirt (https://ovirt.org/blog/2017/09/ovs-dpdk/) This is very interesting feature which can make a huge performance difference in terms of network performance. Just wanted to ask if anyone has tested it in any environment and made any comparison, specially for packet forward (e.g: running Virtual Routers or Virtual Firewalls with virtio) or packet dropping as well. One doubt I have and if someone could clarify is: Should I enable DPDK in the Host any traffic forwarded to any VMs will automatically benefit from this performance gain of DPDK or there additional steps that need to be put in place inside the VM when sharing a physical Network Interface ? Thanks Fernando --------------BDE85DB92A8FD76EF257B82B Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 7bit <html> <head> <meta http-equiv="content-type" content="text/html; charset=utf-8"> </head> <body bgcolor="#FFFFFF" text="#000000"> Hello As some may have seen recently OVS DPDK has been introduced to oVirt (<a class="moz-txt-link-freetext" href="https://ovirt.org/blog/2017/09/ovs-dpdk/">https://ovirt.org/blog/2017/09/ovs-dpdk/</a>). This is very interesting feature which can make a huge performance difference in terms of network performance. Just wanted to ask if anyone has tested it in any environment and made any comparison, specially for packet forward (e.g: running Virtual Routers or Virtual Firewalls with virtio) or packet dropping as well. One doubt I have and if someone could clarify is: Should I enable DPDK in the Host any traffic forwarded to any VMs will automatically benefit from this performance gain of DPDK or there additional steps that need to be put in place inside the VM when sharing a physical Network Interface ? Thanks Fernando </body> </html> --------------BDE85DB92A8FD76EF257B82B--

2 2

Upgrading from Hosted Engine 3.6 to 4.X
by Cam Wright 28 Nov '17

28 Nov '17

Hi there, We're looking to upgrade our hosted engine setup from 3.6 to 4.0 (or 4.1)... We built the 3.6 setup a couple of years ago with Fedora 22 (we wanted the newer-at-the-time kernel 4.4) on the hosts and engine, but when we move to 4.X we'd like to move to EL7 on the engine (as that seems to be the supported version) and to the oVirt Node ISO installer on the hypervisors. We've got only four hosts in our oVirt datacentre, configured in two clusters. Our current idea is to take a backup of the oVirt database using the backup-restore tool, and to take a 'dd' of the virtual disk too, for good measure. Then upgrade the engine to 4.X and confirm that the 3.6 hosts will run, and then finally piecemeal upgrade the hosts to 4.X using the oVirt Node ISO installer. Looking at this page - https://www.ovirt.org/documentation/self-hosted/chap-Maintenance_and_Upgrad… - it seems the 'hosted-engine --upgrade-appliance' path is the best way to do this... but because our hosts are running Fedora instead of EL, I think that makes this option moot to us. Is what I've suggested a valid upgrade path, or is there a more sane way of going about this? -C Cam Wright - Systems and Technical Resource Administrator CUTTINGEDGE / 90 Victoria St, West End, Brisbane, QLD, 4101 T + 61 7 3013 6200 M 0420 827 007 E cwright(a)cuttingedge.com.au | W www.cuttingedge.com.au /SYD /BNE /TYO -- This email is confidential and solely for the use of the intended recipient. If you have received this email in error please notify the author and delete it immediately. This email is not to be distributed without the author's written consent. Unauthorised forwarding, printing, copying or use is strictly prohibited and may be a breach of copyright. Any views expressed in this email are those of the individual sender unless specifically stated to be the views of Cutting Edge Post Pty Ltd (Cutting Edge). Although this email has been sent in the belief that it is virus-free, it is the responsibility of the recipient to ensure that it is virus free. No responsibility is accepted by Cutting Edge for any loss or damage arising in any way from receipt or use of this email. This email may contain legally privileged information and privilege is not waived if you have received this email in error.

3 4

Re: [ovirt-users] VDSM multipath.conf - prevent automatic management of local devices
by Yaniv Kaul 28 Nov '17

28 Nov '17

On Nov 28, 2017 12:24 AM, "Ben Bradley" <listsbb(a)virtx.net> wrote: On 23/11/17 06:46, Maton, Brett wrote: > Might not be quite what you're after but adding > > # RHEV PRIVATE > > To /etc/multipath.conf will stop vdsm from changing the file. > ||| > Hi there. Thanks for the reply. Yes I am aware of that and it seems that's what I will have to do. I have no problem with VDSM managing the file, I just wish it didn't automatically load local storage devices into multipathd. It doesn't. Recent multipath releases do. Does it bother you in any way? Y. I'm still not clear on the purpose of this automatic management though. >From what I can tell there is no difference to hosts/clusters made through this automatic management - i.e. you still have to add storage domains manually in oVirt. Could anyone give any info on the purpose of this auto-management of local storage devices into multipathd in VDSM? Then I will be able to make an informed decision as to the benefit of letting it continue. Thanks, Ben > On 22 November 2017 at 22:42, Ben Bradley <listsbb(a)virtx.net <mailto: > listsbb(a)virtx.net>> wrote: > > Hi All > > I have been running ovirt in a lab environment on CentOS 7 for > several months but have only just got around to really testing things. > I understand that VDSM manages multipath.conf and I understand that > I can make changes to that file and set it to private to prevent > VDSM making further changes. > > I don't mind VDSM managing the file but is it possible to set to > prevent local devices being automatically added to multipathd? > > Many times I have had to flush local devices from multipath when > they are added/removed or re-partitioned or the system is rebooted. > It doesn't even look like oVirt does anything with these devices > once they are setup in multipathd. > > I'm assuming it's the VDSM additions to multipath that are causing > this. Can anyone else confirm this? > > Is there a way to prevent new or local devices being added > automatically? > > Regards > Ben > _______________________________________________ > Users mailing list > Users(a)ovirt.org <mailto:Users@ovirt.org> > http://lists.ovirt.org/mailman/listinfo/users > <http://lists.ovirt.org/mailman/listinfo/users> > > > _______________________________________________ Users mailing list Users(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/users

1 0

oVirt self-hosted Engine [Failed to acquire lock: error -243]
by Terry hey 28 Nov '17

28 Nov '17

Hello all, I installed ovirt self-hosted engine. Unfortunately, the engine VM was suddenly shutdown. And later, it automatically powered on. The engine admin console showed the following error. VM HostedEngine is down with error. Exit message: resource busy: Failed to acquire lock: error -243. I would like to know what is this error message is talking about and how to solve this error. I would like to first thank all of you help me to solve this issue. Regards, Terry

3 3

Auto-starting VMS in a all-in-one / confusion.
by CRiMSON 27 Nov '17

27 Nov '17

I've been digging thru mailing lists and blogs and I'm a bit confused about how you have VMs auto-start after a reboot in a ovirt system that is setup as all-in-one. >From what I can gather this can be achieved via some startup scripts (or a rc.local foo). But there is no setting inside the WebUI that can be set to achieve this? Or have I missed something.

5 6

Error during delete snapshot
by Arthur Melo 27 Nov '17

27 Nov '17

Can someone help me with this error? Failed to delete snapshot '<UNKNOWN>' for VM 'proxy03'. Atenciosamente, Arthur Melo Linux User #302250

2 7

"Enable Discard" for Ceph/Cinder Disks?
by Matthias Leopold 27 Nov '17

27 Nov '17

Hi, according to http://docs.ceph.com/docs/luminous/rbd/qemu-rbd/ the use of Discard/TRIM for Ceph RBD disks is possible. Openstack seems to have implemented it (https://www.sebastien-han.fr/blog/2015/02/02/openstack-and-ceph-rbd-discard/) In oVirt there is no option "Enable Discard" for Cinder Disks (when choosing IDE or VirtIO-SCSI driver), even when i set "report_discard_supported = true" in Cinder. Are there plans for supporting this in the future? Can i use it right now with custom properties (never tried this before)? thx matthias

2 1

Some roles management issues
by Paolo Margara 27 Nov '17

27 Nov '17

Hi list, access policy to my system has changed recently and now I've to manage non admin users allowed to create VMs only on selected resources. Unfortunately this condition was not expected years ago when the system was installed and so all VNic profiles was created with the option "Allow all users to use this Profile", there is a way to edit this settings after a VNic profile was created or I need to create a new one and set all VMs to use that? The same procedure is applicable also for the network used by the hosted engine? I've tried to remove the "VNicProfileUser" role assigned to "Everyone" for a particular VNic profile but by doing only this users continue to see the logical network in the select box of the VM creation form but they cannot use it getting an error when they effectively create a VM clicking OK (that is quite annoying because it seems a configuration problem). I've also noted that the role "DiskProfileUser" was assigned on storage domain object and to some unknown null object for "Everyone", there is some reason why some roles are assigned to null? I could safely remove them? Greetings, Paolo

1 0

Ovirt 4.2pre Moving disks when VM is running
by Arman Khalatyan 27 Nov '17

27 Nov '17

hi, I have some test enviroment with ovirt "4.2.0-0.0.master.20171114071105.gitdfdc401.el7.centos" 2hosts+2NFS-domains During the multiple disk movement between the domains I am getting this warning: Moving disks while the VMs are running.(this is not so scary red as in 4.1.x :) ) What kind of problems can happen during the movement? Thanks, Arman.

2 4

extending cloud Images in oVirt
by Artem Tambovskiy 27 Nov '17

27 Nov '17

I have a question indirectly releted to Ovirt. I need to move one old setup into VM running in oVirt cluster. The VM was based on on Debian 8.9, so I took a Debian cloude image from https://cdimage.debian.org/cdimage/openstack/8.9.8-20171105/ uploaded it into my cluster and attached it to VM. All looks good but ... the disk shows only 2G and indeed ned more disk space. I tried to edit and add more space - and it didnt work. Any ideas how to extend those cloud images? Regards, Artem

3 2

[ANN] oVirt 4.1.8 Third Release Candidate is now available
by Lev Veyde 27 Nov '17

27 Nov '17

The oVirt Project is pleased to announce the availability of the Third Release Candidate of oVirt 4.1.8, as of November 26th, 2017 This update is the eighth in a series of stabilization updates to the 4.1 series. Starting from 4.1.5 oVirt supports libgfapi [5]. Using libgfapi provides a real performance boost for ovirt when using GlusterFS . Due to a known issue [6], using this will break live storage migration. This is expected to be fixed soon. If you do not use live storage migration you can give it a try. Use [7] for more details on how to enable it. This release is available now for: * Red Hat Enterprise Linux 7.4 or later * CentOS Linux (or similar) 7.4 or later This release supports Hypervisor Hosts running: * Red Hat Enterprise Linux 7.4 or later * CentOS Linux (or similar) 7.4 or later * oVirt Node 4.1 See the release notes draft [3] for installation / upgrade instructions and a list of new features and bugs fixed. Notes: - oVirt Appliance is already available - oVirt Live is already available[4] - oVirt Node is already available[4] Additional Resources: * Read more about the oVirt 4.1.8 release highlights: http://www.ovirt.org/release/4.1. <http://www.ovirt.org/release/4.1.8/>8 <http://www.ovirt.org/release/4.1.8/>/ <http://www.ovirt.org/release/4.1.8/> * Get more oVirt Project updates on Twitter: https://twitter.com/ovirt * Check out the latest project news on the oVirt blog: http://www.ovirt.org/blog/ [1] https://www.ovirt.org/community/ [2] https://bugzilla.redhat.com/enter_bug.cgi?classification=oVirt [3] http://www.ovirt.org/release/4.1. <http://www.ovirt.org/release/4.1.8/>8 <http://www.ovirt.org/release/4.1.8/>/ <http://www.ovirt.org/release/4.1.8/> [4] http://resources.ovirt.org/pub/ovirt-4.1-pre/iso/ [5] http://staged-gluster-docs.readthedocs.io/en/release3.7.0beta1/Features/lib… [6] https://bugzilla.redhat.com/show_bug.cgi?id=1306562 [7] http://www.ovirt.org/develop/release-management/features/storage/glusterfs-… -- Lev Veyde Software Engineer, RHCE | RHCVA | MCITP Red Hat Israel <https://www.redhat.com> lev(a)redhat.com | lveyde(a)redhat.com <https://red.ht/sig> TRIED. TESTED. TRUSTED. <https://redhat.com/trusted>

1 0

ovirt 4.1, hosted engine and hyperconvergence
by Gabriel Stein 27 Nov '17

27 Nov '17

Hi! I'm having problems to add new hosts on ovirt web interface and deploying the hosted engine. Everything works until yum try to install the needed packages. Error: Failed to install Host a.b.c. Yum ['qemu-kvm-tools-ev conflicts with 10:qemu-kvm-tools-1.5.3-141.el7_4.2.x86_64', 'qemu-kvm-tools conflicts with 10:qemu-kvm-tools-ev-2.9.0-16.el7_4.8.1.x86_64']. I already tried: - delete all repos and leave just ovirt and centos-base(doesn't works) - yum update(no updates) Some Data: CentOS Linux release 7.4.1708 (Core) oVirt Engine Version: 4.1.7.6-1.el7.centos Any Hints? Thanks! Best Regards, Gabriel Gabriel Stein ------------------------------ Gabriel Ferraz Stein Tel.: +49 (0) 170 2881531

2 1

Re: [ovirt-users] ovirt 4.2 - performance vm
by Sharon Gratch 26 Nov '17

26 Nov '17

+ ovirt-users On Thu, Nov 23, 2017 at 2:31 PM, Mark Hunt <Mark.Hunt(a)arm.com> wrote: > Sharon, > > > All working ok now , thank you. > Great! Thanks for the update. > When is the prod release for rhv 4.2 due to come out do you know ? > RHV 4.2 public beta which is also oVirt 4.2 GA is planned to be released soon. I'm not sure about dates for the prod release so please refer to RHV support for that. > Also I see migration for perf vm's won't be until version 4.3 , is there a > rough timeline on that ? > Still not I think..but again please refer to RHV support for that. Regards, Sharon > > Regards > > > Mark H. > > > > > > ------------------------------ > *From:* Sharon Gratch <sgratch(a)redhat.com> > *Sent:* 15 November 2017 15:04 > *To:* Mark Hunt > *Subject:* Re: ovirt 4.2 - performance vm > > Hi Mark, > > Thanks for the data you sent. > > There was a bug avoiding High Performance VMs with NUMA node pining from > start running (due to Null pointer exception, which is exactly what you > encountered). > This bug was already fixed in latest oVirt engine master versions. > > Can you please sync your engine to latest master version and re-check if > problem is not reproduced anymore? > > > > And regardless to that bug, just a small recommendation regarding the CPU > pinning settings for your high performance VM 'perf' running on host > euhpc-ovirt3.euhpc.arm.com: > For high performance results I recommend you to pin all 8 vCPUs and not > just the first one. > i.e. CPU pinning topology field should be set to something like: > 0#0_1#1_2#2_3#3_4#4_5#5_6#6_7#7 > > And then the NUMA pinning should be set accordingly, i.e. both vNUMAs 0, 1 > should be pinned to host NUMA 0 instead of 0->1 and 1->0 (since CPUs 0-7 > are part of NUMA 0). > > Regards, > Sharon > > > > > On Tue, Nov 14, 2017 at 11:26 AM, Mark Hunt <Mark.Hunt(a)arm.com> wrote: > > Hi Sharon, > > > > Output from lscpu attached > > > Regards > > > Mark H. > > > > > > > > > > ------------------------------ > *From:* Sharon Gratch <sgratch(a)redhat.com> > *Sent:* 13 November 2017 18:10 > *To:* Mark Hunt > *Subject:* Re: ovirt 4.2 - performance vm > > Hi, > > Sorry for the late response. > > It seems that NUMA pinning failed when running VM 'perf' on host: > euhpc-ovirt3.euhpc.arm.com. <http://euhpc-ovirt3.euhpc.arm.com> > Because whenever you cancel the NUMA pinning, the VM runs without any > error on this host. > But when you set the NUMA pinning, the VM failed to run on the host, then > the scheduler tries to rerun the VM on other host in the cluster but there > is no other host and that's why you get the error of "no available hosts". > > So for having a full picture of the problem, can you please send me the > NUMA settings of this host? > For that you just need to login to this host euhpc-ovirt3.euhpc.arm.com and > run from shell: > >> lscpu > > Can you please send me the result of this lscpu command? > > > Thanks, > Sharon > > > On Mon, Oct 30, 2017 at 12:50 PM, Mark Hunt <Mark.Hunt(a)arm.com> wrote: > > Hi Sharon, > > > Yes I did (numa + hp) , please find attached engine.log from the the > ovirt manager vm. One thing I did note was that the first time I pinned the > numa nodes they showed green in the gui , thereafter they were red !? > > > Regards > > > Mark H. > > > > > > ------------------------------ > *From:* Sharon Gratch <sgratch(a)redhat.com> > *Sent:* 24 October 2017 09:07 > *To:* Mark Hunt > *Subject:* Re: ovirt 4.2 - performance vm > > Hi, > > Sorry for the late response. > > Did you set huge-pages and/or NUMA pinning for this high-performance VM? > If so then maybe no host was found with appropriate settings for running > that VM. > Can you please send me the engine log with relevant errors/messages? > > Thanks, > Sharon > > > On Fri, Oct 20, 2017 at 7:39 PM, Mark Hunt <Mark.Hunt(a)arm.com> wrote: > > Hi Sharon, > > > Is there anything not mentioned on the performance vm page that needs to > be set for this to work ? I have tried starting one and all I get back is > there are no available hosts - when there are !? > > > Regards > > > Mark H. > > > > IMPORTANT NOTICE: The contents of this email and any attachments are > confidential and may also be privileged. If you are not the intended > recipient, please notify the sender immediately and do not disclose the > contents to any other person, use it for any purpose, or store or copy the > information in any medium. Thank you. > > > IMPORTANT NOTICE: The contents of this email and any attachments are > confidential and may also be privileged. If you are not the intended > recipient, please notify the sender immediately and do not disclose the > contents to any other person, use it for any purpose, or store or copy the > information in any medium. Thank you. > > > IMPORTANT NOTICE: The contents of this email and any attachments are > confidential and may also be privileged. If you are not the intended > recipient, please notify the sender immediately and do not disclose the > contents to any other person, use it for any purpose, or store or copy the > information in any medium. Thank you. > > > IMPORTANT NOTICE: The contents of this email and any attachments are > confidential and may also be privileged. If you are not the intended > recipient, please notify the sender immediately and do not disclose the > contents to any other person, use it for any purpose, or store or copy the > information in any medium. Thank you. >

1 0

Shared vs Local Storage for Datacenters
by Matt . 26 Nov '17

26 Nov '17

Hi Guys, I'm wondering at the moment what the actual difference is between Shared and Local Storage Datacenters. As far as I can see a Local DC supports NFS as well which makes it a more flexible DC in storage if you ask me, so why do both still exist ? Is it possible in a decent way to change a running Shared Storage DC to a Local one and Keep the NFS shares and add some local mounts for specific hosts ? I hope someone can clearify! Cheers, Matt

4 3

Why Node was rebooting?
by Jonathan Baecker 25 Nov '17

25 Nov '17

This is a multi-part message in MIME format. --------------D1D0407AE4CFA3C37FF62CF4 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Hello community, yesterday evening one of our nodes was rebooted, but I have not found out why. The engine only reports this: 24.11.2017 22:01:43 Storage Pool Manager runs on Host onode-1 (Address: onode-1.worknet.lan). 24.11.2017 21:58:50 Failed to verify Host onode-1 power management. 24.11.2017 21:58:50 Status of host onode-1 was set to Up. 24.11.2017 21:58:41 Successfully refreshed the capabilities of host onode-1. 24.11.2017 21:58:37 VDSM onode-1 command GetCapabilitiesVDS failed: Client close 24.11.2017 21:58:37 VDSM onode-1 command HSMGetAllTasksStatusesVDS failed: Not SPM: () 24.11.2017 21:58:22 Host onode-1 is rebooting. 24.11.2017 21:58:22 Kdump flow is not in progress on host onode-1. 24.11.2017 21:57:51 Host onode-1 is non responsive. 24.11.2017 21:57:51 VM playout was set to the Unknown status. 24.11.2017 21:57:51 VM gogs was set to the Unknown status. 24.11.2017 21:57:51 VM Windows2008 was set to the Unknown status. [...] There is no crash report, and no relevant errors in dmesg. Does the engine send a reboot command to the node, when it gets no responds? Is there any other way to found out why the node was rebooting? The node hangs on a usv and all other servers was running well... In the time, when the reboot was happen, I had a bigger video compression job in one of the VMs, so maybe the CPUs got a bit stressed, but they are not over committed. Regards Jonathan --------------D1D0407AE4CFA3C37FF62CF4 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 7bit <html> <head> <meta http-equiv="content-type" content="text/html; charset=utf-8"> </head> <body text="#000000" bgcolor="#FFFFFF"> Hello community, yesterday evening one of our nodes was rebooted, but I have not found out why. The engine only reports this: <blockquote> <blockquote> 24.11.2017 22:01:43 Storage Pool Manager runs on Host onode-1 (Address: onode-1.worknet.lan). 24.11.2017 21:58:50 Failed to verify Host onode-1 power management. 24.11.2017 21:58:50 Status of host onode-1 was set to Up. 24.11.2017 21:58:41 Successfully refreshed the capabilities of host onode-1. 24.11.2017 21:58:37 VDSM onode-1 command GetCapabilitiesVDS failed: Client close 24.11.2017 21:58:37 VDSM onode-1 command HSMGetAllTasksStatusesVDS failed: Not SPM: () 24.11.2017 21:58:22 Host onode-1 is rebooting. 24.11.2017 21:58:22 Kdump flow is not in progress on host onode-1. 24.11.2017 21:57:51 Host onode-1 is non responsive. 24.11.2017 21:57:51 VM playout was set to the Unknown status. 24.11.2017 21:57:51 VM gogs was set to the Unknown status. 24.11.2017 21:57:51 VM Windows2008 was set to the Unknown status. [...] </blockquote> </blockquote> There is no crash report, and no relevant errors in dmesg. Does the engine send a reboot command to the node, when it gets no responds? Is there any other way to found out why the node was rebooting? The node hangs on a usv and all other servers was running well... In the time, when the reboot was happen, I had a bigger video compression job in one of the VMs, so maybe the CPUs got a bit stressed, but they are not over committed. Regards Jonathan </body> </html> --------------D1D0407AE4CFA3C37FF62CF4--

2 2

Cannot import copied Storage Domain
by Matt . 25 Nov '17

25 Nov '17

Hi Guys, I had a host with a local storage domain which was mounted using NFS and I was not able to deactivate and detach it from the Datacenter so I moved the folder like with content like: 728a7458-e1f9-4a67-8226-668cf3cc546e __DIRECT_IO_TEST__ to a seperate folder and destroyed the old storage domain. Now I want to import the storage domain again which does not work, userrights are ok but I get the message: There is no storage domain under the specified path. Check event pane for more details I also tried to create a new storage domain and put all my images in the images folder but also not working and I don't see my disks or Vm's in oVirt. I have doen such thing in the past and mostly it saved me but now I'm not lucky there. What is the best way to move my images back into oVirt or a new Storage domain ? Maybe someone has an idea here. Thanks! Matt

1 1

oVirt Remote Access
by Emerson Jr 24 Nov '17

24 Nov '17

Hi, I'm having a hard time understanding what happened after I configured oVirt network ("bridge" on the CentOS), it changed the routes metric and now I can't access my oVirt portal from another computer on the network (https:// <IP>/UserPortal). It was working before, I have two Lan interfaces on the network of the server, one of them the IP is bridged to the oVirt, the other one is not, I was using the not bridge IP to access it. Thanks.

2 1

Hosts flapping between non-responive and active
by Matt . 24 Nov '17

24 Nov '17

Hi guys, I have a problem with hosts tha are flapping between NON-responsive and Active. The network is performing well, the VM's are running fine when it happens all the time. I also move a host to a Local Storage DC only without any NFS share attached and even there it happens. In the logs, vdsm.log on the hosts and the engine.log on the engine does not show anything wrong what refers to this behaviour. Where can I look for more information ? My hosts are on the latest 4.1 VDSM versions: Engine: oVirt Engine Version: 4.1.7.6-1.el7.centos Cheers, Matt

2 2