On 29 June 2016 at 18:30, Lindsay Mathieson <lindsay.mathieson at gmail.com> wrote:> Same problem again. VM froze and heal info timed out with "Not able to > fetch volfile from glusterd". I'm going to have to revert to 3.7.11Heal process seems to be stuck at the following: gluster v heal datastore4 info Brick vnb.proxmox.softlog:/tank/vmdata/datastore4 Status: Connected Number of entries: 0 Brick vng.proxmox.softlog:/tank/vmdata/datastore4 <gfid:be318638-e8a0-4c6d-977d-7a937aa84806> - Possibly undergoing heal Status: Connected Number of entries: 1 Brick vna.proxmox.softlog:/tank/vmdata/datastore4 <gfid:be318638-e8a0-4c6d-977d-7a937aa84806> - Possibly undergoing heal Status: Connected Number of entries: 1 I'm on my home now, will be offline for a couple of hours. But for now my cluster is offline. -- Lindsay
Lindsay, Did you see any problems in the setup before you set those options? Also, could you please share glusterd and glfsheal logs before you revert to 3.7.11, so that it can be analyzed? ----- Original Message -----> From: "Lindsay Mathieson" <lindsay.mathieson at gmail.com> > To: "gluster-users" <gluster-users at gluster.org> > Sent: Wednesday, June 29, 2016 2:07:30 PM > Subject: Re: [Gluster-users] 3.7.12 disaster > > On 29 June 2016 at 18:30, Lindsay Mathieson <lindsay.mathieson at gmail.com> > wrote: > > Same problem again. VM froze and heal info timed out with "Not able to > > fetch volfile from glusterd". I'm going to have to revert to 3.7.11 > > > Heal process seems to be stuck at the following: > > gluster v heal datastore4 info > Brick vnb.proxmox.softlog:/tank/vmdata/datastore4 > Status: Connected > Number of entries: 0 > > Brick vng.proxmox.softlog:/tank/vmdata/datastore4 > <gfid:be318638-e8a0-4c6d-977d-7a937aa84806> - Possibly undergoing heal > > Status: Connected > Number of entries: 1 > > Brick vna.proxmox.softlog:/tank/vmdata/datastore4 > <gfid:be318638-e8a0-4c6d-977d-7a937aa84806> - Possibly undergoing heal > > Status: Connected > Number of entries: 1 > > I'm on my home now, will be offline for a couple of hours. But for now > my cluster is offline. > > -- > Lindsay > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://www.gluster.org/mailman/listinfo/gluster-users >-- Thanks, Anuradha.
On Wed, Jun 29, 2016 at 2:07 PM, Lindsay Mathieson < lindsay.mathieson at gmail.com> wrote:> On 29 June 2016 at 18:30, Lindsay Mathieson <lindsay.mathieson at gmail.com> > wrote: > > Same problem again. VM froze and heal info timed out with "Not able to > > fetch volfile from glusterd". I'm going to have to revert to 3.7.11 > > > Heal process seems to be stuck at the following: > > gluster v heal datastore4 info > Brick vnb.proxmox.softlog:/tank/vmdata/datastore4 > Status: Connected > Number of entries: 0 > > Brick vng.proxmox.softlog:/tank/vmdata/datastore4 > <gfid:be318638-e8a0-4c6d-977d-7a937aa84806> - Possibly undergoing heal > > Status: Connected > Number of entries: 1 > > Brick vna.proxmox.softlog:/tank/vmdata/datastore4 > <gfid:be318638-e8a0-4c6d-977d-7a937aa84806> - Possibly undergoing heal > > Status: Connected > Number of entries: 1 > > I'm on my home now, will be offline for a couple of hours. But for now > my cluster is offline. >hi Lindsay, Something doesn't sound right. If you had enabled locking-scheme to granular, 'Possibly undergoing heal' message should never come. I think the problem seems to be something to do with this option in your case. VMs freezing suggest that some processes are still doing old style locking which can block the writes. Can you try resetting the locking-scheme option?> > -- > Lindsay > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://www.gluster.org/mailman/listinfo/gluster-users >-- Pranith -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160629/0e57fcb3/attachment.html>