Description of problem: When running our kernel tier 1 tests, we are noticing large amounts of soft lockup warnings to the point our tests timeout and fail. Only recently we have noticed more and more of these failures. Some examples are: Checking dmesg for specific failures! BUG: soft lockup - CPU#1 stuck for 10s! [rhts-submit-log:32765] BUG: soft lockup - CPU#1 stuck for 10s! [avahi-daemon:2233] BUG: soft lockup - CPU#1 stuck for 10s! [events/1:9] BUG: soft lockup - CPU#0 stuck for 10s! [kjournald:379] End of log. The following links show the failures for kernel-2.6.18-151.el5 http://rhts.redhat.com/cgi-bin/rhts/test_log.cgi?id=8318960 http://rhts.redhat.com/cgi-bin/rhts/test_log.cgi?id=8313998 http://rhts.redhat.com/cgi-bin/rhts/test_log.cgi?id=8318018 http://rhts.redhat.com/cgi-bin/rhts/test_log.cgi?id=8315830 http://rhts.redhat.com/cgi-bin/rhts/test_log.cgi?id=8315632 Look through my testsuite history, kernel-2.6.18-146.el5 seemed to have expected problems of only rpciod soft lockups (which has been normal for a while), kernel-2.6.18-147.el5 failed to installed, kernel-2.6.18-148.el5 had numerous softlockup failures which were _not_ expected. Version-Release number of selected component (if applicable): kernel-2.6.18-151.el5 How reproducible: very Steps to Reproduce: 1. run a kernel tier1 testsuite and notice how only hvm guests have issues 2. 3. Actual results: Expected results: Additional info:
Is this still a problem, or have these messages gone away? I seem to remember some changed in 5.4 that fixed this particular problem, but my memory is fuzzy. DonZ? Chris Lalancette
Things have quieted down recently and I have not seen the volume of softlockups as I did in the past. So if you want to close this as fixed in CURRENTRELEASE or something, I am ok with that. If that problem pops up again, I know where to find you. :-)
Haha. OK, will do, and as always, feel free to open another bug :). Chris Lalancette