Micah wrote:
Kostik Belousov wrote:
On Thu, Oct 26, 2006 at 09:22:43AM -0700, Micah wrote:
Kostik Belousov wrote:
I saw several similar reports.
Please, submit me the output of the "print *mp" in the same frame.
Also, I'm interested in kernel config.
Is the problem reproducible ?
It seems now that any time I compile openoffice the system will
eventually panic. Other disk intensive jobs, like my nightly
photo-album update, may or may not trigger it. I currently have two
dumps for 6.1p10. I thought this might be failing hardware because
the system has worked fine for nearly a year, but the fact that it
panics on the same line of code every time makes me wonder.
I'm very suspicious to the claim of failing hw since trace is the same
all
times (is this true) ? This looks like memory corruption.
They're not identical, see below. I didn't compare the traces until now
since the proximate cause was the same line of code.
First, I would recommend to update to RELENG_6 due to a number of VFS
fixes.
Second, could you set kern.maxvnodes in the /boot/loader.conf ? Check the
value choosen by kernel by "sysctl kern.maxvnodes", and then set it
to the 2/3 of the reported number and reboot.
I'll try the update first by itself to see.
If I not gather any useful info from that action, I most likely
provide you
with debugging patch.
--
Kostik Belousov.
Thanks for the help,
It's been a little over 24 hours since I upgraded to RELENG_6, and I
haven't had a single panic since. I successfully built OO.org and my
nightly scripts ran through completely. I'll throw some heavy loads at
it tomorrow to stress it some more, but for now it looks like whatever
was wrong has been fixed for 6.2.
Thanks,
Micah
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
