[Pacemaker] drbd : returned 8 (master) instead of the expected value: 7 (not running)

[Dominik Klein]

Thomas Mueller wrote:
> hi
> 
> i'm using pacemaker 1.0.3  / hb 2.9.2-sle11rc9 on debian etch. 
> 
> altough everything is working like expected, these warnings pop up every 
> 5 minutes (the reckeck interval):
> 
> Apr 17 07:04:52 ib002 crmd: [31460]: WARN: do_state_transition: 
> Progressed to state S_POLICY_ENGINE after C_TIMER_POPPED
> Apr 17 07:04:52 ib002 pengine: [5973]: info: unpack_rsc_op: r-drbd0-
> db:0_monitor_0 on ib002 returned 8 (master) instead of the expected 
> value: 7 (not running)
> Apr 17 07:04:52 ib002 pengine: [5973]: WARN: unpack_rsc_op: Operation r-
> drbd0-db:0_monitor_0 found resource r-drbd0-db:0 active in master mode on 
> ib002

That's the "probe" monitor op. It "probes" the resource's state before
doing anything to it. Obviously, it expects the resource to be stopped
right after you defined it, but it found it being in master mode, which
usually means you started and promoted drbd outside of the cluster.

Regards
Dominik

> (full loglines here: http://pastebin.com/m1b60ff7a)
> 
> so pengine thinks drbd resource "db" is master on ib002. here ist the 
> output of "drbdadm state db":
> 
> ib002:~# drbdadm state db
> Secondary/Primary
> 
> and pengine thinks it should be "not running". but it should think that 
> it is "started slave". 
> 
> reading the drbd RA i can't see why it would report 8 as exit code.
> 
> is there a way to dynamically start debug log of the DRDD ressource agent 
> to see why it reports 8 as exit code?
> 
> - Thomas