Server becomes unresponsive with many java tasks attempting to run at the same time on RHEL 4.

Environment

• Red Hat Enterprise Linux 4.6
• s390
• IBM WebSphere

Issue

• Server got hung, need the core dump analysed

Resolution

• Update to RHEL 4 update 7 (kernel 2.6.9-78 or higher)

Root Cause

• The system is hung in the epoll_wait state, as per This content is not included.this bug

Diagnostic Steps

• 'sys' command in crash shows high load average ( > 500)

• 'ps' command in crash shows hundreds of running Java processes

• bt -a shows the active process has sys_epoll_wait at the top of the stack trace:

     #0 [8897fe00] sys_epoll_wait at c3542
     #1 [8897feb8] sysc_noemu at 2f13e
  

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.