[
https://issues.apache.org/jira/browse/QPID-961?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12640543#action_12640543
]
Marnie McCormack commented on QPID-961:
---------------------------------------
Looked extensively (and tested, arrgh) on Solaris and Linus at ps, pgrep, pkill
and kill syntax. End result is that some versions of Solaris always give you a
limited char set of the process name, making relying on the existing process
name not viable.
After much pain, agreed the simplest (yeuch) solution for this is to introduce
a check for a PNAME variable after the vm -server option, adding into qpid-run.
> qpid.stopall script no longer runs correctly under Solaris
> ----------------------------------------------------------
>
> Key: QPID-961
> URL: https://issues.apache.org/jira/browse/QPID-961
> Project: Qpid
> Issue Type: Bug
> Components: Java Broker
> Affects Versions: M2.1
> Reporter: Martin Ritchie
> Assignee: Aidan Skinner
> Fix For: M4
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> Summary:
> The problem we have is this command line:
> pids=`ps -ef |grep $USER | grep $PROGRAM | grep -v grep | awk '{print $2}'`
> Which now as a result in adding XX values to our startup script shows:
> ritchiem 10188 7608 0 14:44:40 pts/34 0:08 java -server
> -XX:+UseConcMarkSweepGC -XX:+HeapDumpOnOutOfMemoryError -Xmx1024m
> Which doesn't exactly allow us to identify it as Qpid.
> We should if possible move the -DQPID_HOME option to be at the start or there
> may be a better set of options to ps for Solaris.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
