On February 3, 2020 5:06:16 AM GMT+02:00, Christian Reiss
<[email protected]> wrote:
>Hey,
>
>they're all in this form:
>
>[root@node03:[..]/images/6113f079-fd28-4165-a807-61bb7625cd48] # l
>total 49G
>drwxr-xr-x. 2 vdsm kvm 8.0K Jan 29 23:02 .
>drwxr-xr-x. 46 vdsm kvm 8.0K Feb 2 23:18 ..
>-rw-rw----. 1 vdsm kvm 50G Jan 29 02:02
>83f7942f-c74e-4bc4-a816-09988e7ab308
>-rw-rw----. 1 vdsm kvm 1.0M Jan 23 12:16
>83f7942f-c74e-4bc4-a816-09988e7ab308.lease
>-rw-r--r--. 1 vdsm kvm 323 Jan 29 23:02
>83f7942f-c74e-4bc4-a816-09988e7ab308.meta
>-rw-rw----. 1 vdsm kvm 20G Feb 2 21:42
>f72a4a62-b280-4bdf-9570-96d4b6577d89
>-rw-rw----. 1 vdsm kvm 1.0M Jan 29 23:02
>f72a4a62-b280-4bdf-9570-96d4b6577d89.lease
>-rw-r--r--. 1 vdsm kvm 251 Jan 29 23:02
>f72a4a62-b280-4bdf-9570-96d4b6577d89.meta
>
>Looks good (enough) to me.
>
>
>On 03/02/2020 04:01, Darrell Budic wrote:
>> Check the contents of these directories:
>>
>>
>[root@node03:/rhev/data-center/mnt/glusterSD/node01.dc-dus.dalason.net
>>
><http://node01.dc-dus.dalason.net>:_ssd__storage/fec2eb5e-21b5-496b-9ea5-f718b2cb5556/images]
>
>> # l
>> total 345K
>> drwxr-xr-x. 46 vdsm kvm 8.0K Feb 2 23:18 .
>> drwxr-xr-x. 5 vdsm kvm 64 Feb 3 00:31 ..
>> drwxr-xr-x. 2 vdsm kvm 8.0K Jan 17 15:54
>> 0b21c949-7133-4b34-b909-a6660ae12800
>> drwxr-xr-x. 2 vdsm kvm 165 Feb 3 01:48
>> 0dde79ab-d773-4d23-b397-7c39371ccc60
>> drwxr-xr-x. 2 vdsm kvm 8.0K Jan 17 09:49
>> 1347d489-012b-40fc-acb5-d00a9ea133a4
>> drwxr-xr-x. 2 vdsm kvm 8.0K Jan 22 15:04
>> 1ccc4db6-f47d-4474-b0fa-a0c1eddb0fa7
>> drwxr-xr-x. 2 vdsm kvm 8.0K Jan 21 16:28
>> 22cab044-a26d-4266-9af7-a6408eaf140c
>> drwxr-xr-x. 2 vdsm kvm 8.0K Jan 30 06:03
>> 288d061a-6c6c-4536-a594-3bede63c0654
>> drwxr-xr-x. 2 vdsm kvm 8.0K Jan 9 16:46
>> 40c51753-1533-45ab-b9de-2c51d8a18370
>>
>> and what version of Ovirt are you running? This looks a bit like a
>> libvirt change/bug that changed ownership on the actual disk image to
>
>> root.root on shutdown/migrations, preventing later start attempts.
>>
>> This may help if that’s the case:
>> chown -R vdsm.kvm
>> /rhev/data-center/mnt/glusterSD/node01.dc-dus.dalason.net
>>
><http://node01.dc-dus.dalason.net>:_ssd__storage/fec2eb5e-21b5-496b-9ea5-f718b2cb5556/images
>>
>>> On Feb 2, 2020, at 8:54 PM, Christian Reiss
><[email protected]
>>> <mailto:[email protected]>> wrote:
>>>
>>> Hey,
>>>
>>> it was _while_ placing the host _into_ maintenance, to be precise.
>>> I restarted the volumes and even each machine and the entire cluster
>
>>> to no avail.
>>>
>>> I am currently migrating the disk images out of ovirt into
>openvz/kvm
>>> to get them running. The copied disk images are flawless and
>working.
>>>
>>>
>>> On 03/02/2020 03:28, Jayme wrote:
>>>> I checked my HCI cluster and those permissions seem to match what
>I'm
>>>> seeing. Since there's no VMs running currently have you tried
>>>> restarting the gluster volumes as well as the glusterd service? I'm
>
>>>> not sure what would have caused this with one host placed in
>maintenance.
>>>
>>> --
>>> with kind regards,
>>> mit freundlichen Gruessen,
>>>
>>> Christian Reiss
>>> _______________________________________________
>>> Users mailing list -- [email protected] <mailto:[email protected]>
>>> To unsubscribe send an email to [email protected]
>>> <mailto:[email protected]>
>>> 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/[email protected]/message/FOFNQO4FLJAIUB3EEXFBZ3HGWFI3ACU5/
>>
Deja view for me.
Enable brick trace log (for sgort time or you run out of space) and check if
acl is the reason.
What is your gluster version ?
Did you test VM power off & power on after the last gluster upgrade ?
If it is acl, you have 3 options (not valid for 7.1 & 7.2):
1. Mount with acl enabled
mount -t glusterfs -o acl brick1:/volume1 /mnt
And run a dummy setfacl:
find /mnt -exec setfacl -u:root:rw {} \;
2. Kill gluster processes and start the volume with 'force' option:
gluster volume start <volume> force (or something like that.
3. Maybe a downgrade, yet 'm not in productive environment and that could be
different for you.
Best Regards,
Strahil Nikolov
_______________________________________________
Users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
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/[email protected]/message/4KPA7TTWZ3P4ELBCRKLHGUOGYXPSG5RZ/
