Description of problem: xose 7533 44.9 0.1 1496 276 pts/0 R 23:02 1:43 ./lucifer xose 7538 34.4 0.1 1492 276 pts/0 R 23:03 0:59 ./lucifer xose 7544 26.2 0.1 1504 380 pts/0 R 23:04 0:30 ./lucifer xose 7559 25.1 0.1 1488 276 pts/0 R 23:05 0:15 ./lucifer CPU is more than 100% -> impossible but top, mpstat, vmstat are ok Version-Release number of selected component (if applicable): procps-2.0.11-6 How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
This is probably not a procps bug, but just the usual poor process accounting that Linux does. However, that should show up in top too. When you say you don't see it in top, do you mean the total cpu usage or in the actual process listings?
ps shows impossible CPU values. But with top, mpstat or vmstat I get correct CPU values. This is a very clear ps bug
I forgot it. top shows _correct values_ in the total cpu usage _and_ in the actual process listings.
Is this fixed by version 3.1.15 or later versions?
I will see it....
Upstream procps says that this is a kernel problem.
ok. I'm going to close it, now RHL is dead. -thanks-