Hi
Node 0
UID PID PPID C STIME TTY TIME CMD
db2poc 7072 7070 6 15:55 pts/0 00:00:00 db2sysc 0
root 7073 7072 0 15:55 pts/0 00:00:00 db2ckpwd 0
root 7074 7072 0 15:55 pts/0 00:00:00 db2ckpwd 0
root 7075 7072 0 15:55 pts/0 00:00:00 db2ckpwd 0
Thats the output from db2_local_ps
On Fri, Aug 14, 2009 at 3:29 PM, Dejan Muhamedagic <[email protected]>wrote:
> Hi,
>
> On Fri, Aug 14, 2009 at 02:42:32PM +0200, Timothy Carr wrote:
> > Heh,
> >
> > thanks for putting me on the correct path. You are 100% correct.
>
> The other poster actually figured out what was going on.
>
> > the pscount code which is being returned is a value of 5. This is
> > unfortunately too much. The amount of processes returned from DB2 9.5 are
> > actually 4 and therefore the db2 resource script will fail because the "
> > test $pscount -ge 5 " looks for the amount of db2 processess greater and
> > equal to 5. Set the value to 3 and you will be set :)
>
> I guess you meant 4?
>
> I just wonder whether that would introduce a regression.
> Can you show a sample of the output of db2_local_ps?
>
> Thanks,
>
> Dejan
>
> > Thanks
> >
> >
> > On Fri, Aug 14, 2009 at 1:54 PM, Dejan Muhamedagic <[email protected]
> >wrote:
> >
> > > Hi,
> > >
> > > On Fri, Aug 14, 2009 at 01:32:22PM +0200, Timothy Carr wrote:
> > > > Hi All,
> > > >
> > > > I've setup linux HA and its working pretty well with my disks being
> > > migrated
> > > > when i need it to and VIP working nicely.
> > > > The problem I seem to have is getting DB2 to startup with the OCF
> > > resource
> > > > agent.
> > > >
> > > >
> > > > crmd[6031]: 2009/08/14_13:06:51 info: do_lrm_rsc_op: Performing
> > > > op=db2poc_monitor_0
> key=5:27162:7:596c511b-85e0-40b6-b1df-2ef28e41fa94)
> > > > lrmd[6028]: 2009/08/14_13:06:51 info: rsc:db2poc: monitor
> > > > crmd[6031]: 2009/08/14_13:06:52 info: process_lrm_event: LRM
> operation
> > > > db2poc_monitor_0 (call=47, rc=7) complete
> > > > crmd[6031]: 2009/08/14_13:06:55 info: do_lrm_rsc_op: Performing
> > > > op=db2poc_start_0
> key=19:27162:0:596c511b-85e0-40b6-b1df-2ef28e41fa94)
> > > > lrmd[6028]: 2009/08/14_13:06:55 info: rsc:db2poc: start
> > > > db2[20759]: 2009/08/14_13:07:02 INFO: DB2 UDB instance db2poc
> started:
> > > > 08/14/2009 13:07:02 0 0 SQL1063N DB2START processing was
> > > > successful.
> > > > SQL1063N DB2START processing was successful.
> > > > db2[20759]: 2009/08/14_13:07:02 ERROR: DB2 UDB instance db2poc not
> > > > active!
> > > > crmd[6031]: 2009/08/14_13:07:02 info: process_lrm_event: LRM
> operation
> > > > db2poc_start_0 (call=48, rc=1) complete
> > > > crmd[6031]: 2009/08/14_13:07:03 info: do_lrm_rsc_op: Performing
> > > > op=db2poc_stop_0 key=1:27163:0:596c511b-85e0-40b6-b1df-2ef28e41fa94)
> > > > lrmd[6028]: 2009/08/14_13:07:03 info: rsc:db2poc: stop
> > > > db2[21059]: 2009/08/14_13:07:05 INFO: DB2 UDB instance db2poc
> stopped:
> > > > 08/14/2009 13:07:05 0 0 SQL1064N DB2STOP processing was
> > > successful.
> > > > SQL1064N DB2STOP processing was successful.
> > > > db2[21059]: 2009/08/14_13:07:06 INFO: DB20000I The TERMINATE
> command
> > > > completed successfully.
> > > > db2[21059]: 2009/08/14_13:07:07 INFO: ipclean: Removing DB2 engine
> and
> > > > client's IPC resources for db2poc.
> > > > crmd[6031]: 2009/08/14_13:07:07 info: process_lrm_event: LRM
> operation
> > > > db2poc_stop_0 (call=49, rc=0) complete
> > > > cib[6027]: 2009/08/14_13:16:44 info: cib_stats: Processed 32
> operations
> > > > (312.00us average, 0% utilization) in the last 10min
> > > >
> > > >
> > > > The folllowing is an extract from the ha-log file. It starts up the
> > > instance
> > > > but then somehow detects that the instance is not active then shuts
> down
> > > DB2
> > > > again.
> > >
> > > It looks like the db2 resource agent can't handle newer db2
> > > releases. In particular, the monitor operation fails since db2
> > > started using threads. Somebody reported it two-three weeks ago
> > > on this list. The problem is here:
> > >
> > > pscount=`runasdb2 $db2bin/db2_local_ps | cut -c9- | grep ' db2[^ ]' |
> wc
> > > -l`
> > > test $pscount -ge 5
> > >
> > > We need to find out a db2 release running and then implement
> > > a proper test.
> > >
> > > Unfortunately I don't have a db2 instance for testing.
> > >
> > > Thanks,
> > >
> > > Dejan
> > >
> > > >
> > > > --
> > > > Timothy Carr
> > > > Technical
> > > > Gtalk: [email protected]
> > > > Skype: timothy.carr.foxtrail
> > > > _______________________________________________
> > > > 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
> > >
> >
> >
> >
> > --
> > Timothy Carr
> > Technical Specialist
> > University of Cape Town
> > Cell: +27834572568
> > Fax: +27865472190
> > Gtalk: [email protected]
> > Skype: timothy.carr.foxtrail
> > _______________________________________________
> > 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
>
--
Timothy Carr
Technical Specialist
University of Cape Town
Cell: +27834572568
Fax: +27865472190
Gtalk: [email protected]
Skype: timothy.carr.foxtrail
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
