On Sun, 2004-02-22 at 17:02, xyzzy@xxxxxxxxxx wrote: > Look at bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=109497 which > has been open since November and is ending up with the consensus that it is a > problem with the Fedora stock kernels.... People with this bug report that > if they run 2.4.24 from kernel.org, they have uptimes of over a week without > problems. And how is Red Hat "not caring" w.r.t. this BZ entry? I looked at it and the last "Red Hat activity" is davej's post from 2004-01-20. Granted this is one month ago, but we're all in FC2 test frenzy since quite a while and have heaps of stuff which needs to be done before the next test release (others way much more than I). Mind you that Fedora is a Community Project (with a capital "C"), "community" being first and foremost about giving, not taking. This is one point where also people with non-RH addresses can come into play ;-) -- configure a serial console from another computer, log the traces/oopses/whatever, trace them back to where it fails and compare the FC1 kernel to kernel.org's one where it differs in that regard. This shouldn't be exactly hard for a kernel developer as you are ;-). > I was also leaning towards trying the 2.6 kernel but that is a major hassle > for me if I can't get X up at all due to the seg fault I had from updating > the Intel graphics driver. Really, try out the 2.6 kernel, it is hugely different from 2.4 stuff... Running the following command might help you finding what the Intel binary crap^Wonly drivers messed up in XFree86 space: rpm -V $(rpm -qa XFree86\*) Also look out for libGL*.so* files not owned by XFree86 and remove them: locate '*/libGL*.so*' rpm -qf .... (<- result from above cmd here) Nils -- Nils Philippsen / Red Hat / nphilipp@xxxxxxxxxx "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -- B. Franklin, 1759 PGP fingerprint: C4A8 9474 5C4C ADE3 2B8F 656D 47D8 9B65 6951 3011
Attachment:
signature.asc
Description: This is a digitally signed message part
