Hi! Bumping an old thread, because there’s now activity around this bug. The github issue is https://github.com/gluster/glusterfs/issues/2492 We just hit this bug after an update from GlusterFS 7.x to 9.4. We did not see this in our test environment, so we did the update, but the bug is still there. Apparently the fix should be https://github.com/gluster/glusterfs/pull/2509 which should get backported to 9.x.
We worked around this issue by identifying the server with the bug and restarting the GlusterFS processes on it. On an EL/CentOS/Fedora-based system there was one small thing that surprised me, maybe this will help others. There’s the service /usr/lib/systemd/system/glusterfsd.service which does not really start anything, just runs /bin/true, but when stopped, will kill the brick processes on the server. If you try doing “systemctl stop glusterfsd” but you have not started the service (even though starting it does nothing), systemd will not do anything. If you first start the service and then stop it, systemd will actually run the ExecStop command. Best regards, Ville-Pekka ________ Community Meeting Calendar: Schedule - Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC Bridge: https://meet.google.com/cpu-eiue-hvk Gluster-users mailing list [email protected] https://lists.gluster.org/mailman/listinfo/gluster-users
