Description of problem: ======================= As part of the fix for bz: 1401817, snapshot creation will fail in a racy scenario where barrier brick op will fail while bricks are coming up, due to this snapshot creation will fail too. Version-Release number of selected component (if applicable): ============================================================= glusterfs-3.8.4-6.el7rhgs.x86_64 How reproducible: ================= Its a race Steps to Reproduce: =================== 1. Create 2*2 distributed replicate volume 2. enabled scheduler, 3. scheduled snapshot every one minute, while snapshot are getting created bring bricks offline/online.
Hi Avra, I wanted to check if there is any workaround for this issue.
Hi Bhavana, The doc text looks good. But there is no workaround to this problem. The snapshot create will fail and needs to be retried.
Since barrier xlator enable or disable was failed because of the graph initialization , then I think it is an expected failure. As long as it fail properly then it won't create any problem. Unless there is any un-recoverable problem with the failure, we don't need to keep this bug open.