[mythtv-users] OT: FedoraCore3 + NFS + XFS + Software Raid = EIP?

Blammo blammo.doh at gmail.com
Thu Jun 30 22:11:02 EDT 2005


On 6/30/05, Nick <knowledgejunkie at gmail.com> wrote:
> On 7/1/05, Blammo <blammo.doh at gmail.com> wrote:
> > I know this isn't a kernel mailing list, however, given that this list
> > has a large number of people who combine these features, I thought I'd
> > give it a shot.
> >
> > I have a backend running (at the moment) 2.6.12.1 w/8k stacks. I've
> > been fighting this issue however, since about 2.6.7. going from 4k
> > stacks to 8k stacks helped, but didn't remove the problem.
> >
> > Occasionally, specific to heavy IO, I'll have the machine go into a
> > state of limbo. It will either Hardlock (no keyboard activity) or go
> > into Limbo. The times it doesn't hard lock, I've been able to track it
> > back to NFS.
> >
> >  I've replaced motherboard, CPU, RAM, Video, ethernet, HD's, power
> > supplies, and kernel versions, chased IRQ's, etc, so I'm fairly sure
> > it's a bug.
> >
> > Anyone want to field a guess?
> 
> Have you tried with any other distros, like CentOS or WBEL, which are
> RH Enterprise based (may contain specific patches not in FC)?
> 
> Assuming the disks are fsck'd without problem and your swap is not
> corrupted, and this problem is reproducible on a *completely*
> different system (non-Via), I'd def. take it to linux.kernel if you
> haven't already with traces at the ready.

WBEL / Centos based off RHEL4 don't have XFS support (and I haven't
found any way to get it in), and those based off RHEL3 are on a 2.4
kernel, which leaves me out-of-luck for DVB.

The disks all test clean. I've low-leveled them all.


More information about the mythtv-users mailing list