Created attachment 588028 [details] list of RPMs installed at time of problem Description of problem: Having updated everything and rebooted yesterday (see attached RPM list), I started getting kernel panics at random times while doing nothing in particular (except using the browser, which I guess could mean anything). This was not happening in the week or so before yesterday. I don't know how to send in a report other than by attaching pictures of the monitor (please let me know if there's a better way). Smolt profile: http://www.smolts.org/client/show_all/pub_a311dc29-a9b9-4b06-bfee-ed29946befd9 Version-Release number of selected component (if applicable): kernel-3.3.7-1.fc17.x86_64 for the first two pictures. I then booted to kernel-3.3.6-3.fc17.x86_64 and still got another oops (last picture). How reproducible: Not sure how to reproduce, although it seemed like it might be related to Google Chrome re-opening all the same tabs to look at each time. I had 4 panics in a day of light use. Might be able to reproduce by opening all those tabs again? Additional info: Might be related to setting the following to bring sanity to my dual monitors: gsettings set org.gnome.shell.overrides workspaces-only-on-primary false c.f. http://gregcor.com/2011/05/07/fix-dual-monitors-in-gnome-3-aka-my-workspaces-are-broken/
Created attachment 588029 [details] first oops under kernel-3.3.7-1.fc17.x86_64
Created attachment 588030 [details] second oops under kernel-3.3.7-1.fc17.x86_64
Created attachment 588031 [details] oops under kernel-3.3.6-3.fc17.x86_64 too
The F16 oops is irrelevant since you have virtualbox modules loaded. They don't appear to be loaded in the F17 oopses, but the one you have marked as "first" already has the B flag set in taint, so something else bugged before that one. Is it available in /var/log/messages? The oops corresponds to this code: static inline int put_page_testzero(struct page *page) { VM_BUG_ON(atomic_read(&page->_count) == 0); <<<------ return atomic_dec_and_test(&page->_count); } Which is weird. You might want to run memtest on this machine overnight.
Thanks for taking a look. The earlier kernel is from the F17 beta, but you're right, I have virtualbox modules loaded with that (might as well get rid of it, won't run with my hardware anyway). I will look in /var/log/messages and do a memory check when I get a chance.
You were right on target; memtest uncovered hardware problems. Thanks again for looking at this, and sorry for the red herring.