i think i figured it out now, and it works stable now for a week:

the jboss resource was missing this option: *pstring*
="^/usr/jdk/latest/bin/java.-Dprogram.name=run.sh.*10.100.102.105"

i did not define it at first because it was stated its optional but when i
did read the jboss run script in resources it seemed that its pretty often
used.

anyway thanks.

E



On Wed, Jan 19, 2011 at 13:36, Erik Dobák <[email protected]> wrote:

> thank you i did not ask for exact causes just for hints as i am new to
> heartbeat.
>
> ad 1) yes i know
> ad 2) i will not argue with you about this one
> ad your hints) i am the only admin of those machines no one has access i
> suspect now a virtual machine backup to screw up things. will disable it and
> lets see.
>
> cheers
>
> E
>
> ps: there is a shortage of crystal balls worldwide, somebody should do
> something about this ;)
>
>
> On Wed, Jan 19, 2011 at 13:27, Andrew Beekhof <[email protected]> wrote:
>
>> On Wed, Jan 19, 2011 at 12:14 PM, Erik Dobák <[email protected]>
>> wrote:
>> > yes but why did it time out?
>>
>> you're asking me why your unique instance of jboss, the one none of us
>> have ever seen, took too long to shutdown?
>>
>> > the monitor checks the jboss index page and i could access it manualy
>> > without problem.
>>
>> 1) monitor != stop
>> 2) results now have no baring on past or future results
>>
>> Maybe someone ran a fork-bomb at the time, or initiated a backup, or...
>>
>> Sorry, we don't have crystal balls.
>>
>> >
>> > did try to crm resources cleanup and reprobe but no success.
>> >
>> > i restarted now both nodes and it is running fine, lets see what will
>> > happen.
>> >
>> > E
>> >
>> >
>> > On Wed, Jan 19, 2011 at 12:02, Andrew Beekhof <[email protected]>
>> wrote:
>> >
>> >> On Wed, Jan 19, 2011 at 11:02 AM, Erik Dobák <[email protected]>
>> wrote:
>> >> > i have a cluster running on 1 node the resources are active. on the
>> other
>> >> > they are passive.
>> >> > when i did status i got only STARTED for both resources.
>> >> >
>> >> > but over night seems something went wrong see below.
>> >> > the strange thing is that both resources the ipaddr2 and jboss are
>> >> running
>> >> > correctly but heartbeat does not think so.
>> >> >
>> >> > any idea why?
>> >>
>> >> Because the monitor and stop operations "Timed Out" perhaps?
>> >>
>> >> >
>> >> >
>> >> > [root@lc-cl1 ~]# crm_mon -1
>> >> > ============
>> >> > Last updated: Wed Jan 19 10:37:35 2011
>> >> > Stack: Heartbeat
>> >> > Current DC: lc-cl2 (ecde9589-6940-49a5-a45f-a79574dfde33) - partition
>> >> with
>> >> > quorum
>> >> > Version: 1.0.10-da7075976b5ff0bee71074385f8fd02f296ec8a3
>> >> > 2 Nodes configured, unknown expected votes
>> >> > 1 Resources configured.
>> >> > ============
>> >> >
>> >> > Online: [ lc-cl2 ]
>> >> > OFFLINE: [ lc-cl1 ]
>> >> >
>> >> >  Resource Group: bamcluster
>> >> >     ipaddr2    (ocf::heartbeat:IPaddr2):       Started lc-cl2 FAILED
>> >> >     lcbam      (ocf::heartbeat:jboss): Started lc-cl2 (unmanaged)
>> FAILED
>> >> >
>> >> > Failed actions:
>> >> >    lcbam_stop_0 (node=lc-cl2, call=8, rc=-2, status=Timed Out):
>> unknown
>> >> > exec error
>> >> >    ipaddr2_monitor_10000 (node=lc-cl2, call=11, rc=-2, status=Timed
>> Out):
>> >> > unknown exec error
>> >> > _______________________________________________
>> >> > Linux-HA mailing list
>> >> > [email protected]
>> >> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
>> >> > See also: http://linux-ha.org/ReportingProblems
>> >> >
>> >> _______________________________________________
>> >> Linux-HA mailing list
>> >> [email protected]
>> >> http://lists.linux-ha.org/mailman/listinfo/linux-ha
>> >> See also: http://linux-ha.org/ReportingProblems
>> >>
>> > _______________________________________________
>> > Linux-HA mailing list
>> > [email protected]
>> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
>> > See also: http://linux-ha.org/ReportingProblems
>> _______________________________________________
>> Linux-HA mailing list
>> [email protected]
>> http://lists.linux-ha.org/mailman/listinfo/linux-ha
>> See also: http://linux-ha.org/ReportingProblems
>>
>
>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to