[Pacemaker] Reason for automatic migration after one node rebooted?

Andrew J. Caines A.J.Caines at halplant.com
Thu Feb 6 13:00:46 EST 2014

I'm new to Pacemaker and am working on my RTFM+STFW efforts, but have an
urgent need to find the cause of an automatic service migration after a
node was rebooted and would appreciate assistance.

This is a two node active/passive cluster with a single service - NFS.
One node, nfs01, has a preference rule in its configuration[1]:

  location cli-prefer-nfs_services nfs_services \
	rule $id="cli-prefer-rule-nfs_services" inf: #uname eq nfs01

The sequence of events was:

1. nfs01 and nfs02 both up and running with NFS service on nfs02,

2. Manual migration of NFS services to nfs01 [crm resource migrate
   nfs_services nfs01] fails[2],

4. nf01 is rebooted the next day,

5. NFS service immediately migrates to nfs01[3][4].

Was this because

A. The preference rule only
B. The migration attempt only
C. The preference rule and the migration attempt
D. Other?

If more detail is needed, then I'll be happy to provide it.

[1] http://pastebin.com/raw.php?i=3ThD1uM7
[2] http://pastebin.com/raw.php?i=5F9142SF
[3] http://pastebin.com/raw.php?i=LA4E0vUS
[4] http://pastebin.com/raw.php?i=6BpB5L4u

-Andrew J. Caines-   Unix Systems Engineer   A.J.Caines at halplant.com
FreeBSD/Linux/Solaris, Web/Mail/Proxy/...   http://halplant.com:2001/
  "Machines take me by surprise with great frequency" - Alan Turing

More information about the Pacemaker mailing list