10GB disk created on glusterfs storage has only 4096B in vm

Hi, I have another spooky issue. CentOS 8.2, oVirt 4.4.1.10-1 I am trying to create vm on my HCI cluster (which in fact has only two nodes now, the third will be added later). I have created new 10GB disk on gluster storage (it behaves the same way when I chose prealocated or thin provision disk). Everything looks good (I was surprised how fast was this disk created) but when I run the vm then disk has only 4096 bytes. see http://mirror.slu.cz/tmp/ovirt_small_disk.png disk creation part of engine.log is here https://paste.slu.cz/?0e33e800609898d8#CEC2maoe7zk9b1zmnHaAENjaAnpiyzzGzPFSe... the same part in vdsm.log on SPM host https://paste.slu.cz/?e89fdce800c39b28#3UWw7dXxQAmZysMSKp5NbeTWMNDghZVh2NLEm... where could be problem? Did anybody experience this behavior? Cheers, Jiri

I had a similar problem, after i migrate the disk from a "Storage Domain" to a second "Storage Domain" the disk size was different and i could not start the vm anymore. The Ovirt error was : "Unable to get volume size for domain" Somehow gluster has some errors when i migrate disks and the disk have different size. What i did was to disable performance.stat-prefetch for the gluster volume(I read it has some bugs if is enabled in gluster 7.x) Also try to look at Disk Snapshots, see if you have a snapshot that does not exist. In my case i had this snapshot problem after migration and i had to delete the snapshot from the database and set the main image disk id as default in the vm. Maybe in you case is different. Emy

hello, On 7/29/20 6:33 PM, shadow emy wrote:
I had a similar problem, after i migrate the disk from a "Storage Domain" to a second "Storage Domain" the disk size was different and i could not start the vm anymore. The Ovirt error was : "Unable to get volume size for domain" Somehow gluster has some errors when i migrate disks and the disk have different size. What i did was to disable performance.stat-prefetch for the gluster volume(I read it has some bugs if is enabled in gluster 7.x)
thanks for reply. In my case it loks like I have not "performance.stat-prefetch" enabled... [root@ovirt-hci01 ~]# gluster volume info vms Volume Name: vms Type: Replicate Volume ID: ba5fd3b8-0704-4462-8257-94b77a1222c4 Status: Started Snapshot Count: 0 Number of Bricks: 1 x 2 = 2 Transport-type: tcp Bricks: Brick1: 10.0.4.13:/gluster_bricks/vms/vms Brick2: 10.0.4.11:/gluster_bricks/vms/vms Options Reconfigured: performance.client-io-threads: off cluster.eager-lock: enable performance.io-cache: off performance.read-ahead: off performance.quick-read: off user.cifs: off network.ping-timeout: 30 network.remote-dio: off performance.strict-o-direct: on performance.low-prio-threads: 32 features.shard: on storage.owner-gid: 36 storage.owner-uid: 36 transport.address-family: inet storage.fips-mode-rchecksum: on nfs.disable: on cluster.self-heal-daemon: enable ...well, If it is not a default setting... so I set it off... [root@ovirt-hci01 ~]# gluster volume set vms performance.stat-prefetch off volume set: success ...and unbeliavable, it works now! Is it bug in glusterfs 7.6-1.el8 (link?) or something else? Should not be this setting default one?
Also try to look at Disk Snapshots, see if you have a snapshot that does not exist. In my case i had this snapshot problem after migration and i had to delete the snapshot from the database and set the main image disk id as default in the vm.
I have no snapshots of this vm. I just created new vm and try to install it from ISO.
Maybe in you case is different.
Well, it was exactly this case, thanks! Cheers, Jiri
Emy _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/WO7S2ZFTSRPRY2...

On Wed, Jul 29, 2020 at 7:13 PM Jiří Sléžka <jiri.slezka@slu.cz> wrote:
thanks for reply. In my case it loks like I have not "performance.stat-prefetch" enabled...
[root@ovirt-hci01 ~]# gluster volume info vms
Volume Name: vms Type: Replicate Volume ID: ba5fd3b8-0704-4462-8257-94b77a1222c4 Status: Started Snapshot Count: 0 Number of Bricks: 1 x 2 = 2 Transport-type: tcp Bricks: Brick1: 10.0.4.13:/gluster_bricks/vms/vms Brick2: 10.0.4.11:/gluster_bricks/vms/vms Options Reconfigured: performance.client-io-threads: off cluster.eager-lock: enable performance.io-cache: off performance.read-ahead: off performance.quick-read: off user.cifs: off network.ping-timeout: 30 network.remote-dio: off performance.strict-o-direct: on performance.low-prio-threads: 32 features.shard: on storage.owner-gid: 36 storage.owner-uid: 36 transport.address-family: inet storage.fips-mode-rchecksum: on nfs.disable: on cluster.self-heal-daemon: enable
BTW: you can run this command to see all options settings, default ones and not for the vms volume: gluster volume get vms all

On 7/29/20 7:37 PM, Gianluca Cecchi wrote:
On Wed, Jul 29, 2020 at 7:13 PM Jiří Sléžka <jiri.slezka@slu.cz <mailto:jiri.slezka@slu.cz>> wrote:
thanks for reply. In my case it loks like I have not "performance.stat-prefetch" enabled...
[root@ovirt-hci01 ~]# gluster volume info vms
Volume Name: vms Type: Replicate Volume ID: ba5fd3b8-0704-4462-8257-94b77a1222c4 Status: Started Snapshot Count: 0 Number of Bricks: 1 x 2 = 2 Transport-type: tcp Bricks: Brick1: 10.0.4.13:/gluster_bricks/vms/vms Brick2: 10.0.4.11:/gluster_bricks/vms/vms Options Reconfigured: performance.client-io-threads: off cluster.eager-lock: enable performance.io-cache: off performance.read-ahead: off performance.quick-read: off user.cifs: off network.ping-timeout: 30 network.remote-dio: off performance.strict-o-direct: on performance.low-prio-threads: 32 features.shard: on storage.owner-gid: 36 storage.owner-uid: 36 transport.address-family: inet storage.fips-mode-rchecksum: on nfs.disable: on cluster.self-heal-daemon: enable
BTW: you can run this command to see all options settings, default ones and not for the vms volume:
gluster volume get vms all
good point! Cheers, Jiri

Good that is ok for you now. As Gianluca told you the command to see all the gluster volume settings is : gluster volume get vol_name all The previous command you used : gluster volume info vol_name will list only modified settings from default and not all the settings. The output shows only volume "Options Reconfigured". Regarding if this is an important bug that affects gluster 7.6(or other versions) and should be disabled by default, i dont know for sure. I am just using ovirt as a hyperconvergence tool. Maybe someone from the ovirt development team can answer that.

I have been using 7.6 (and rewntly migrated to 7.7) on my oVirt 4.3.10 withkut any issues so far. Are you sure that it's not oVirt 4.4 specific ? Best Regards, Strahil Nikolov На 30 юли 2020 г. 15:03:17 GMT+03:00, shadow emy <shadow.emy1@gmail.com> написа:
Good that is ok for you now. As Gianluca told you the command to see all the gluster volume settings is : gluster volume get vol_name all
The previous command you used : gluster volume info vol_name will list only modified settings from default and not all the settings. The output shows only volume "Options Reconfigured".
Regarding if this is an important bug that affects gluster 7.6(or other versions) and should be disabled by default, i dont know for sure. I am just using ovirt as a hyperconvergence tool. Maybe someone from the ovirt development team can answer that. _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/EFVT6AMOI4IQJK...

Damn, those thick fingers... На 30 юли 2020 г. 23:12:00 GMT+03:00, Strahil Nikolov <hunter86_bg@yahoo.com> написа:
I have been using 7.6 (and rewntly migrated to 7.7) on my oVirt 4.3.10 withkut any issues so far.
Are you sure that it's not oVirt 4.4 specific ?
Best Regards, Strahil Nikolov
На 30 юли 2020 г. 15:03:17 GMT+03:00, shadow emy <shadow.emy1@gmail.com> написа:
Good that is ok for you now. As Gianluca told you the command to see all the gluster volume settings is : gluster volume get vol_name all
The previous command you used : gluster volume info vol_name will list only modified settings from default and not all the settings. The output shows only volume "Options Reconfigured".
Regarding if this is an important bug that affects gluster 7.6(or other versions) and should be disabled by default, i dont know for sure. I am just using ovirt as a hyperconvergence tool. Maybe someone from the ovirt development team can answer that. _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/EFVT6AMOI4IQJK...
participants (4)
-
Gianluca Cecchi
-
Jiří Sléžka
-
shadow emy
-
Strahil Nikolov