Dual processor machine with FreeBSD 5.2 hangs or reboots regulary

John Baldwin jhb at FreeBSD.org
Thu Feb 5 11:30:32 PST 2004


On Thursday 05 February 2004 09:47 am, Ain Tepp wrote:
> On Wed, 4 Feb 2004, John Baldwin wrote:
> > On Wednesday 04 February 2004 05:02 am, Ain Tepp wrote:
> > > Hello!
> > >
> > > I have problem with following configuration:
> > >
> > > FreeBSD 5.2
> > > Dual Xeon 2, 4 GHz
> > > Gigabyte 8IPXDR-E motherboard
> > > 2 x Intel e1000 (em, integrated) net interface
> > > 1 x LSI MegaRAID (amr) controller
> > > 4 x 512 MB 266 Reg ECC Kingston memory
> > >
> > > Problem description:
> > > Server  hangs regulary with 5.1. After upgrade 5.2 problems continue,
> > > but
>
> a
>
> > > little. If random device removed from kernel configuration, server
> > > doesn't hang any more, but after 3 days uptime reboots.
> > > Messages in console are like "supervisor read, page fault", server
> > > trying restart, but disk syncing and processor halting didn't succeed.
> > > One time error associates with httpd process, other errors seems to be
> > > em0 problem.
> > >
> > > Done:
> > > Hardware tests, parts change.
> > >
> > > I have a doubt that kernel SMP and em0 don't work affiliate.
> >
> > Unfortunately, it is hard to track this problem down without more
>
> information.
>
> > Are you able to build a debug kernel with DDB and get a traceback that
> > you can send here?
>
> Thank You for answer!
>
> Sorry, but physical access in server is very complicated, it be located in
> isp server room.
> if i can doing anything without trapping in debugger reboot time, let's
> advice me.

You can try building a debug kernel and turning on crashdumps.  Then using gdb 
on the kernel.debug and vmcore file to get the kernel panic messages and a 
traceback.

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org



More information about the freebsd-smp mailing list