W/o this patch, puppet sends notification
to Service['ceph'] unconditionally. As a
result, Service['ceph'] is being restarted,
followed by all deployed OSD restart on the
entire node.
This is a problem since all the related
Ceph OSD services has been already started
by `ceph deploy osd activate` and this notify
with restart only creates huge load
(especially in cases when OSD is added to
long-existing cluster under the load) and may
lead to unstable errors when particular OSD
services fail to start.
The solution is to trigger ceph service restart
only on configuration updates.
Reviewed: https:/ /review. openstack. org/207761 /git.openstack. org/cgit/ stackforge/ fuel-library/ commit/ ?id=141f6f7ed70 39644b5e451cc52 1f2fc492ae6e81
Committed: https:/
Submitter: Jenkins
Branch: master
commit 141f6f7ed703964 4b5e451cc521f2f c492ae6e81
Author: Bogdan Dobrelya <email address hidden>
Date: Fri Jul 31 09:47:37 2015 +0200
Restart ceph service only on conf change
W/o this patch, puppet sends notification
to Service['ceph'] unconditionally. As a
result, Service['ceph'] is being restarted,
followed by all deployed OSD restart on the
entire node.
This is a problem since all the related
Ceph OSD services has been already started
by `ceph deploy osd activate` and this notify
with restart only creates huge load
(especially in cases when OSD is added to
long-existing cluster under the load) and may
lead to unstable errors when particular OSD
services fail to start.
The solution is to trigger ceph service restart
only on configuration updates.
Closes-bug: #1479880
Change-Id: Ib60436d5037904 1a6a1368b0fa170 b9b58cc484c
Signed-off-by: Bogdan Dobrelya <email address hidden>