ALERT: Some images may not load properly within the Knowledge Base Article. If you see a broken image, please right-click and select 'Open image in a new tab'. We apologize for this inconvenience.

Vmstat processes accumulating on Sun Solaris systems where Applications Manager is installed


Article ID: 88858


Updated On:


CA Automic Applications Manager (AM)


Vmstat processes accumulating on Sun Solaris systems where Applications Manager is installed


Release: AAMOS499000-8.0-Automic Applications Manager-OS400 Agent


Applicable to Applications Manager Versions: v7.1, v8.0


Vmstat processes can hang onsome Sun Solaris systems where Applications Manager is running causing them to accumulate which can cause the CPU usageon that server to spike.


Applications Manager uses vmstat to monitorCPU usage on the Master and Agent machines in order to load balance betweenAgents and to stop job processing on Agents if the CPU becomes overloaded. On some Sun Solaris machines these vmstatprocesses can hang because they do not end properly and can accumulate overtime.


This issue was addressed with therelease of AppWorx v 71 HP2 and Applications Manger v8.0 (Bug 25143). However, the code changes made for this bug do not always succeed inkilling the vmstat processes on some Sun Solaris machines, which is due to aproblem with the Solaris operating system itself which causes the processes tohang. Contact Sun in order to address the issue or use the workaround outlined below.


You can periodically manuallykill any vmstat processes that have accumulated or you can create a script on yourSolaris agents that will run the following command as the Applications Manageruser and schedule it to run as needed:

kill -9 `ps -ef|grepvmstat|grep -v vmstat|awk '{ print $2 }'`

Be sure to test the script in a non-production environment prior to moving it into production.