[Linux-HA] the return code of failing start action

Junko IKEDA ikedaj at intellilink.co.jp
Wed Oct 10 04:05:39 MDT 2007


> The last time start of prmDummy occured in lrmd was here:
> 
> tengine[24482]: 2007/10/09_13:42:59 WARN: action_timer_callback: Timer
> popped (abort_level=0, complete=false)
> tengine[24482]: 2007/10/09_13:42:59 WARN: print_elem: Action missed its
> timeout[Action 4]: In-flight (id: prmDummy_start_0, loc: prec370e,
priority:
> 0)
> lrmd[24141]: 2007/10/09_13:42:59 WARN: prmDummy:start process (PID 24564)
> timed out (try 1).  Killing with signal SIGTERM (15).
> lrmd[24141]: 2007/10/09_13:42:59 info: RA output: (prmDummy:start:stderr)
> Terminated
> 
> Dummy[24564][24582]: 2007/10/09_13:42:59 INFO: They use TERM to bring us
down.
> No such luck.
> lrmd[24141]: 2007/10/09_13:43:04 WARN: prmDummy:start process (PID 24564)
> timed out (try 2).  Killing with signal SIGKILL (9).
> lrmd[24141]: 2007/10/09_13:43:04 WARN: Exiting prmDummy:start process
24564
> killed by signal 9 [SIGKILL - Kill, unblockable].
> lrmd[24141]: 2007/10/09_13:43:04 WARN: operation start[3] on
> ocf::Dummy::prmDummy for client 24144, its parameters:
> CRM_meta_id=[opDummyStart] delay=[1] CRM_meta_timeout=[10000]
> crm_feature_set=[2.0] CRM_meta_name=[start] : pid [24564] timed out

I have modified Dummy script to cause a timed out.
so, It's not surprising that RA would return rc=-1.
I had checked this value, and created the split brain on purpose.
after a few minutes, I fixed the split brain.
The return code was -2 at that time somehow.
I expected that the value was still -1 after recovering split brain.
Where is "-2" from?

Thanks,
Junko





More information about the Linux-HA mailing list