Last updated on NOVEMBER 20, 2014
Applies to:Enterprise Manager Base Platform - Version 188.8.131.52.0 and later
IBM AIX on POWER Systems (64-bit)
Deployed EM 12c agent and server rebooted. System logs details says that emdprocstats.pl consuming was 10gb of system cache+free mem.
"<agent_base>/core/184.108.40.206.0/sysman/admin/scripts/emdprocstats.pl" collects agent health metric for every 15 min. Script takes the agent PID as an arguments:
System Admin feedback on crash:
The actual PID of interest is 47513654. The PID is being trimmed in Perceive to 7513654.
The start time of for PID 47513654 is 1/29/2013 12:37 CST.
PID 47513654 appears to have been consuming on the order of 10 GB just prior to server crash.
The memory consumption by this process would have driven the file system cache to flush to accommodate increased demand for process memory.
Once both the free memory and file system cache dropped below acceptable levels, physical memory was exhausted and the system began to page out to virtual memory (aka swap).
When swap filled up, the system crashed.
The full process name for PID 47513654 is:
/opt/oracle/product/220.127.116.11.AGT/core/18.104.22.168.0/perl/bin/perl /opt/oracle/product/22.214.171.124.AGT/core/126.96.36.199.0/sysman/admin/scripts/emdprocstats.pl 38404174
Sign In with your My Oracle Support account
Don't have a My Oracle Support account? Click to get started
My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms