Red Hat Bugzilla – Bug 50636
pe7150 unresponsive while running cerberus
Last modified: 2007-04-18 12:35:23 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2-2 i686; en-US; m18) Gecko/20010131 Netscape6/6.01 Description of problem: While running cerberus 9-20 on a pe7150 w/ beta3 installed, 24GB ram, 2GB swap, perc3qc raid 0 container, bios x15. Cerberus stops timer within 15 minutes, change of virtual console takes 30 seconds, top takes over 30 minutes to update. All swap space in use. All but 8MB ram in use. Alt-sysreq is available. No disk IO occuring. How reproducible: Always Steps to Reproduce: 1.Install fairfax beta3 on pe7150 2.Run cerberus 9-20 and note performance, responsiveness, and timer stopping 3. Actual Results: System all but hangs Expected Results: Top should update frequently, virtual console available immediately, disk IO should be evident, cerberus timer should continue. Additional info:
Behaviour did not occur on a pe7150 w/ beta 3 installed, bios X15, 2GB ram, 2GB swap.
Does this behavior happen with the 2.4.7 kernel, or with more swap added?
Yes and Yes. Kernel 2.4.7-0.12.0smp, with 24GB ram, 12GB swap. Heartbeat stopped after 3 hours. Machine was still up but same unresponsiveness. Still had sysreq access. Could not ctrl-c cerberus to kill it.
This defect is considered MUST-FIX for Fairfax.
Newburn ran for over 3 days with kernel 2.4.9-0.18smp and 32GB ram and 2GB swap. Machine retained responsiveness and newburn could be stopped and started again without problem.