OC4J 10.1.3 Performance Slows Down When the Load Increases by More Concurrent Users Accessing the Application.
(Doc ID 731010.1)
Last updated on DECEMBER 27, 2020
Applies to:Oracle Containers for J2EE - Version 10.1.3.0.0 to 10.1.3.5.0 [Release AS10gR3]
Information in this document applies to any platform.
When the load increases as more users are accessing the application, OC4J performance slows down leading to different timeout errors. These errors can be found in following files:
- OPMN log file : default_group~<oc4j instance>~default_group~1.log
java.lang.IllegalStateException: IOException: An established connection was aborted by the software in your host machine
- HTTP Server: error:_log.xxx
[Wed Jun 18 07:18:58 2008] [error][client <ip>] [ecid:<ecid>] mod_oc4j: Failed to find a failover oc4j process for session request for destination: <oc4j instance> (no island or jgroup).
[Wed Jun 18 07:26:32 2008] [warn] [client <ip>] oc4j_read_client_data timed out
[Wed Jun 18 07:48:32 2008] [warn] [client <ip>] oc4j_read_client_data timed out
Furthermore following information indicates this problem:
- OC4J thread dump shows large number (200+) of threads idle and waiting on socket read even when there are only 20 or fewer number of requests
- Running "netstat -an|grep <ajp port>" shows large number of sockets on "TIME_WAIT"
Also setting the OC4J startup option -Dajp.keepalive=true for the specific oc4j instance does not help.
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document