Generic Data Services restarts for no reason.
Hi,
I have configured Generic Data Service to moniter JBoss instance running on port 8080 on my two cluster node.
Problem is,
I have a script which starts jboss and terminates and the pid is released.
Jboss daemon runs with another pid.
Since the original PID is released after the script ends, sun cluster is assuming that the process has died and is restarting the resource.
How to write a startup script for sun cluster, so that it does the activity, stays alive but not use cpu cycles.
Thanks for the help in advance.
Message was edited by:
ballu121
[616 byte] By [
ballu121] at [2007-11-26 10:38:21]
# 1
If you are refering to the configuration you described at
http://forum.sun.com/jive/thread.jspa?threadID=108635&tstart=0
then from what you write your issue is with setting "-x Child_mon_level=0".
Any specific reason for that?
To cite the SUNW.gds manpage:
snip
Child_mon_level (integer)
This property provides control over the processes that
are monitored through the Process Monitor Facility
(PMF). This property denotes the level to which the
forked children processes are monitored. Omitting this
property or setting this property to the default value
is the same as omitting the -C option for pmfadm(1M):
all children (and their descendents) are monitored.
snip
so as you noticed, your initial pid disappears, since you set the value to 0 any child processes are not traced by pmf - so the last pid disappears from pmf and consequently pmf decides the service is gone.
I would recommend either leaving the value to default or setting it to a level that fits to the jboss behaviour.
Greets
Thorsten
