Bug 1641673 - pacemaker service is disabled after creating nfs-ganesha cluster.
Summary: pacemaker service is disabled after creating nfs-ganesha cluster.
Keywords:
Status: CLOSED DUPLICATE of bug 1637564
Alias: None
Product: Red Hat Gluster Storage
Classification: Red Hat Storage
Component: common-ha
Version: rhgs-3.4
Hardware: x86_64
OS: Linux
unspecified
medium
Target Milestone: ---
: ---
Assignee: Kaleb KEITHLEY
QA Contact: Manisha Saini
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2018-10-22 13:45 UTC by Kaleb KEITHLEY
Modified: 2019-07-10 06:37 UTC (History)
9 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of: 1441055
Environment:
Last Closed: 2018-11-22 04:28:57 UTC
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2017:2774 0 None None None 2018-10-22 13:45:12 UTC

Comment 7 Kaleb KEITHLEY 2018-11-02 12:03:25 UTC 
https://code.engineering.redhat.com/gerrit/154376
Comment 11 Kaleb KEITHLEY 2018-11-05 13:03:43 UTC 
At some point the `pcs cluster setup ...` (invoked during cluster setup) started to `systemctl disable pacemaker`. We want pacemaker to be enabled so that it starts automatically after a reboot.

(It doesn't seem to have always done this AFAIK. There is a track record of pcs/pacemaker/corosync changing behaviour from release to release.)

A `systemctl enable pacemaker` was added (circa rhgs-3.2 perhaps) to the ganesha-ha.sh at the end of the setup processing to attempt to address this. That worked until rhgs-3.4 (and even appears to have worked when tested by QE during 3.4).

But the customer is still seeing it. I suspect there is a race condition where this could manifest on a busy system.

Now there is an --enable option to `pcs cluster setup` that will ensure that even with a race condition that pacemaker will always be enabled  at the end of the cluster setup.
Comment 13 Jiffin 2018-11-22 04:28:57 UTC 

*** This bug has been marked as a duplicate of bug 1637564 ***
Note You need to log in before you can comment on or make changes to this bug.