[Bugs] [Bug 1600941] [geo-rep]: geo-replication scheduler is failing due to unsuccessful umount

bugzilla at redhat.com bugzilla at redhat.com
Fri Jul 13 12:45:36 UTC 2018


https://bugzilla.redhat.com/show_bug.cgi?id=1600941



--- Comment #3 from Kotresh HR <khiremat at redhat.com> ---
>From Csaba's analysis in comment 3, The master volume mount done by scheduler
script is being used by geo-rep worker. That's really strange. When it's
analysed deeper, it's found out that it's related to private mount namespace
being used by worker. The monitor spawns the worker with private mount
namespace. Once the workers are up, the mounts done outside worker are not
propagated to worker and vice versa. But the global mounts done prior to the
worker spawn are propagated to the worker. Even though it's propagated, it
should not block the original process which had mounted the global mount to
umount and clean up the directory.

But that's the behaviour in the system that is reproduced (Centos 7).
So mitigating the issue by mounting it after the worker process is up.

[root at dhcp42-119 ~]# uname -a
Linux dhcp42-119.lab.eng.blr.redhat.com 3.10.0-862.3.2.el7.x86_64 #1 SMP Tue
May 15 18:22:15 EDT 2018 x86_64 x86_64 x86_64 GNU/Linux

-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=PvnlmEJFUu&a=cc_unsubscribe


More information about the Bugs mailing list