Gluster users - Mar 2020 - Image File Owner change Situation. (root:root)

Hello @All,

This has happened to us for the first time and only on One VM.

I believe it happened with the switch from Fuse to "LibgfApi" in
Ovirt.

I was using LibgfApiSupported=True on 4.2.8 . I upgraded to 4.3.8 and did NOT
restart all of my VMs (30+)

but only some VMs, no problem. Eventually I noticed that LibgfApiSupported=False
and reset it to True.

The VM was Running WindowsServer2016 and we did a Cold-Reboot (not VM Restart).
It did not come back online
due to "Invalid Volume" which was eventually due to the IMAGEs (Boot
and Data) being  user:group=root but
not the meta,lease or directory. Nor any other VM have/had this problem but they
were (if at all) completely
stopped and restarted.

I am looking for another VM that has not yet been restarted to test this theory.
I thought this would be
interesting for others looking into this problem.

(I will ask my Colleague next week what he means with Cold-Reboot vs
Warm-reboot....)

Stay Healthy.

Cheers,

Robert O'Kane



-- 
Robert O'Kane
Systems Administrator
Kunsthochschule f?r Medien K?ln
Peter-Welter-Platz 2
50676 K?ln

fon: +49(221)20189-223
fax: +49(221)20189-49223

Hi Robert,

there were serveral issues with ownership in ovirt, for example see;
https://bugzilla.redhat.com/show_bug.cgi?id=1666795
Maybe you're encountering these issues during the upgrade process. Also if
you're using gluster as backend storage, there might be some permission
issues in the 6.7 or below branches. not sure about the newest version's.

Best Olaf

Op vr 13 mrt. 2020 om 16:55 schreef Robert O'Kane <okane at khm.de>:
> Hello @All,
>
> This has happened to us for the first time and only on One VM.
>
> I believe it happened with the switch from Fuse to "LibgfApi" in
Ovirt.
>
> I was using LibgfApiSupported=True on 4.2.8 . I upgraded to 4.3.8 and did
> NOT restart all of my VMs (30+)
>
> but only some VMs, no problem. Eventually I noticed that
> LibgfApiSupported=False  and reset it to True.
>
> The VM was Running WindowsServer2016 and we did a Cold-Reboot (not VM
> Restart). It did not come back online
> due to "Invalid Volume" which was eventually due to the IMAGEs
(Boot and
> Data) being  user:group=root but
> not the meta,lease or directory. Nor any other VM have/had this problem
> but they were (if at all) completely
> stopped and restarted.
>
> I am looking for another VM that has not yet been restarted to test this
> theory. I thought this would be
> interesting for others looking into this problem.
>
> (I will ask my Colleague next week what he means with Cold-Reboot vs
> Warm-reboot....)
>
> Stay Healthy.
>
> Cheers,
>
> Robert O'Kane
>
>
>
> --
> Robert O'Kane
> Systems Administrator
> Kunsthochschule f?r Medien K?ln
> Peter-Welter-Platz 2
> 50676 K?ln
>
> fon: +49(221)20189-223
> fax: +49(221)20189-49223
> ________
>
>
>
> Community Meeting Calendar:
>
> Schedule -
> Every Tuesday at 14:30 IST / 09:00 UTC
> Bridge: https://bluejeans.com/441850968
>
> Gluster-users mailing list
> Gluster-users at gluster.org
> https://lists.gluster.org/mailman/listinfo/gluster-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.gluster.org/pipermail/gluster-users/attachments/20200313/502fa344/attachment.html>