On 12/01/2014 09:43 PM, Emmanuel Dreyfus wrote: > On Mon, Dec 01, 2014 at 09:42:09PM +0530, Ravishankar N wrote: >> I see the mgmt locks spurious-failure-fix patches referenced above have been merged but I am still able to hit it on master. >> In case some motivated soul wants to have a look, here is the log: http://build.gluster.org/job/rackspace-regression-2GB-triggered/2961/consoleFull > > FWIW, it also strikes quite often on NetBSD. I have not looked at it > since it is obviously OS independent. Its on my radar, I am in progress of analysing it. The last patch set was on the clean up part of the test cases, I felt the changes could have solved it, but I am afraid it didn't. I tried to reproduce it multiple times in my local set up but couldn't. Initial analysis finds out that one of glusterd node got disconnect when multiple volume set transactions were in progress, reason is still unknown. I will keep you posted once I find any significant details. ~Atin. > > _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx http://supercolony.gluster.org/mailman/listinfo/gluster-devel
