3 Replies Latest reply: Jan 3, 2013 11:30 AM by Smitha-Oracle RSS

    shared pool for 12c agent and OMS

      Hi Everyone.
      As per documentation, the heap size can be changed by adding the Xmx and Xms entries in the server startup arguments section of the WLS admin console. My question is when i do a ps -ef | grep java i see below entry
      Middleware/jdk16/jdk/bin/java -server -Xms256m -Xmx1740m -XX:MaxPermSize=768M. so look like the shared pool max is 1.7 GB. But im not able to find this configuration Can someone tell me where this might have been configured?
        • 1. Re: shared pool for 12c agent and OMS
          The OMS setting is defined in the ../gc_inst/user_projects/domains/GCDomain/bin/startEMServer.sh file:

          if [ "${SERVER_NAME}" != "EMGC_ADMINSERVER" ] ; then
          USER_MEM_ARGS="-Xms256m -Xmx1740m -XX:MaxPermSize=768M -XX:-DoEscapeAnalysis -
          if [ "${JAVA_VENDOR}" = "Sun" ] ; then
          if [ "${PRODUCTION_MODE}" = "" ] ; then
          USER_MEM_ARGS="-Xms256m -Xmx1740m -XX:MaxPermSize=768M -XX:-DoEscapeAnalys
          is -XX:+UseCodeCacheFlushing -XX:CompileThreshold=8000 -XX:PermSize=128m -verbos
          e:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps"
          This file contains the env settings and java options used for OMS startup.

          Reference: Enterprise Manager Performance: How to get More Information When OMS Re-starts due to "Out of Memory" Condition? (Doc ID 794165.1)
          • 2. Re: shared pool for 12c agent and OMS
            Thanks for the reply. So if i wanted to change this i would add the arguments in the startup section of the admin console for the admin server, and also the managed servers which will overwrite this setting. Am i correct?
            • 3. Re: shared pool for 12c agent and OMS
              1. If you need to change the value for the OMS the startEMServer.sh script itself needs to be updated as this script overrides all configuration in all other locations.
              Steps for doing this are in the earlier mentioned Doc ID 794165.1.

              2. There is no recommendation for agent heap size based on the number of targets. If the heap size is less, you would ideally see that the agent is crashing due to out-of-memory' errors in emagent.nohup file.