Il 07/02/20 14:51, Strahil Nikolov ha scritto: > On February 7, 2020 10:30:19 AM GMT+02:00, Christian Reiss <email@xxxxxxxxxxxxxxxxxx> wrote: >> Hey, >> >> the ACL issue did not occur during an upgrade. I did upgrade the broken >> >> cluster to get rid of the error. The version was oVirt node 4.3.7 with >> the shipped gluster version. I upgraded to 4.3.8 with Gluster 6.7 and >> let's see how production ready this really is. >> >> -Chris. >> >> On 07/02/2020 08:46, Paolo Margara wrote: >>> Hi, >>> >>> this is interesting, this happen always with gluster 6.6 or only in >>> certain cases? >>> >>> I ask this because I have two ovirt clusters with gluster, both with >>> gluster v6.6, in one case I've upgraded from 6.5 to 6.6 as Strahil, >> and >>> I haven't hit this bug. >>> >>> When upgrading my clusters I follow exactly the steps reported by >>> Strahil in the bug report, some other things should be different. >>> >>> >>> Greetings, >>> >>> Paolo >>> >>> Il 06/02/20 23:30, Christian Reiss ha scritto: >>>> Hey, >>>> >>>> I hit this bug, too. With disastrous results. >>>> I second this post. >>>> >>>> >>>> On 06.02.2020 19:59, Strahil Nikolov wrote: >>>>> Hello List, >>>>> >>>>> Recently I had upgraded my oVirt + Gluster (v6.5 -> v6.6) and I >> hit >>>>> an ACL bug , which forced me to upgrade to v7.0 >>>>> >>>>> Another oVirt user also hit the bug when upgrading to v6.7 and he >>>>> had to rebuild his gluster cluster. >>>>> >>>>> Sadly, the fun doesn't stop here. Last week I have tried upgrading >> to >>>>> v7.2 and again the ACL bug hit me. Downgrading to v7.1 doesn't >> help >>>>> - so I downgraded to 7.0 and everything is operational. >>>>> >>>>> The bug report for the last issue: >>>>> https://bugzilla.redhat.com/show_bug.cgi?id=1797099 >>>>> >>>>> I have 2 questions : >>>>> 1. What is gluster's ACL evaluating/checking ? Is there an option >> to >>>>> force gluster not to support ACL at all ? >>>>> >>>>> 2. Are you aware of the issue? That bug was supposed to be fixed a >>>>> long time ago, yet the facts speak different. >>>>> >>>>> Thanks for reading this long post. >>>>> >>>>> Best Regards, >>>>> Strahil Nikolov >>>>> ________ >>>>> >>>>> Community Meeting Calendar: >>>>> >>>>> APAC Schedule - >>>>> Every 2nd and 4th Tuesday at 11:30 AM IST >>>>> Bridge: https://bluejeans.com/441850968 >>>>> >>>>> NA/EMEA Schedule - >>>>> Every 1st and 3rd Tuesday at 01:00 PM EDT >>>>> Bridge: https://bluejeans.com/441850968 >>>>> >>>>> Gluster-users mailing list >>>>> Gluster-users@xxxxxxxxxxx >>>>> https://lists.gluster.org/mailman/listinfo/gluster-users > Hi Paolo, > > I'm not sure. > Did you set the checkmark to stop gluster, when you set the node in maintenance ? > > Best Regards, > Strahil Nikolov Hi Strahil, no, I stop it manually after I put node in maintenance. For processes that does not stop I kill them manually, there was a procedure for gluster's online upgrade on the documentation but I don't find it now. I always reboot the node when upgrade finished. Greetings, Paolo ________ Community Meeting Calendar: APAC Schedule - Every 2nd and 4th Tuesday at 11:30 AM IST Bridge: https://bluejeans.com/441850968 NA/EMEA Schedule - Every 1st and 3rd Tuesday at 01:00 PM EDT Bridge: https://bluejeans.com/441850968 Gluster-users mailing list Gluster-users@xxxxxxxxxxx https://lists.gluster.org/mailman/listinfo/gluster-users
