We kill jboss by kill -9 on the parent id. We get the parent id at startup like this (snippet from run.sh):
exec "$JAVA" -ms512m -mx1536m $JAVA_OPTS \
-classpath "$JBOSS_CLASSPATH" \
org.jboss.Main "$@" < /dev/null > $jboss_stdout 2>&1 &
echo $! > /usr/local/nick/run/jboss-3.0.6.pid
furthermore, pstree only shows one java process group.
$ pstree -l
init-+-bdflush
|-crond
|-java---java---162*[java]
|-jserver
|-keventd
|-3*[kjournald]
|-klogd
|-ksoftirqd_CPU0
|-ksoftirqd_CPU1
|-kswapd
|-kupdated
|-mdrecoveryd
|-migration_CPU0
|-migration_CPU1
|-6*[mingetty]
|-ntpd
|-portmap
|-scsi_eh_0
|-scsi_eh_1
|-snmpd
|-sshd-+-8*[sshd---sshd]
| `-sshd---sshd---bash---pstree
|-syslogd
`-ypbind---ypbind---2*[ypbind]
If there WERE some badly killed java processes hanging on to the clustering ports, shouldn't I see some java.net.BindExceptions or such like?
A ghost in the machine;-(
Also when I restart a node I get a nice update on the state of the cluster - what nodes are present and such. Is there a way to get this info without restarting a node?
regards,
Nick
Sacha Labourey wrote:
Are you sure you don't have a badly killed jboss intance still running in
the background?
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]] On Behalf Of Nick Mills
Sent: mardi, 18. février 2003 10:16
To: [EMAIL PROTECTED]
Subject: [JBoss-user] multiple instances of same node machine in cluster
Dear clusterers,
Looking at the server.log of my jboss-3.0.6 server and noticed multiple instances of the server named ap01 in the cluster. Any ideas on how the same server node got to be in the same cluster 4 times? And what effect would this have on the running cluster?
here's the server.log output...
2003-02-18 17:20:59,178 DEBUG [org.jboss.ha.framework.server.ClusterPartition] Starting ClusterParti
tion: DefaultPartition
2003-02-18 17:20:59,179 INFO [org.jboss.ha.framework.server.ClusterPartition] Connecting to channel
2003-02-18 17:20:59,203 INFO [STDOUT]
-------------------------------------------------------
GMS: address is ap01:32876
-------------------------------------------------------
2003-02-18 17:20:59,331 INFO [org.jboss.ha.framework.server.HAPartitionImpl.DefaultPartition] New c
luster view: 7 ([ap03:34203, ap04:34361, ap01:32870, ap01:32872, ap01:32874, ap01:32876])
2003-02-18 17:20:59,332 DEBUG [org.jboss.ha.framework.server.HAPartitionImpl.DefaultPartition] ViewA
ccepted: initial members set
...
I don't know if this is relevant but prior to noticing the above in the logs I had trouble restarting the server ap01. On startup log output just stopped at the
"GMS: address is ap01:32876"
part of the cluster initialization. I ran the default jboss configuration once then was able to successfully start my clustered configuration. I am not sure why this happened either, so if anybody has any ideas?
we are running Redhat 7.3 and the sun j2sdk1.4.1_01.
thanks for any info.
cheers
Nick
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf _______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-user
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-user
------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ JBoss-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-user
