System hangs while running benchmark test with Timekeeper installed
Issue
- We use the UnixBench toolkit to benchmark servers in our environment
http://code.google.com/p/byte-unixbench/One of the tests, a process spawn test will consistently hang the system when run with a parallelism of more than 20. The only way to restore the system is with a power reset. This did not happen under Red Hat 5.x or 6.3. I would like to submit a crash dump to pinpoint where this is occurring and see if there is a fix or workaround to prevent user-level processes from completely hanging the server.
Environment
- Red Hat Enterprise Linux 6.4
- timekeeper-6.3.2-0.noarch
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.