Hi Adam -
There were some fixes that went into CVS around the end of the year that
may well fix your problem. Unfortunately, neither of the 1.0.2 kernels
have these fixes. If you can, I'd suggest testing either the 2.4.17
snapshot patch, or doing a fresh CVS checkout.
-Eric
On Thu, 17 Jan 2002, Adam Hupp wrote:
> We have been experiencing frequent hangs with our Redhat 7.2 system
> running the 2.4.9 RH7.2 XFS 1.0.2 kernel rpm. There is nothing in the
> logs, and the machine usually stops dead in it's tracks. No console
> access, no ping, nothing. One time we still had NFS access while the
> rest of the machine (console, ssh, etc ) was inaccessable. I seem to be
> able to trigger it by running lots of fs load, but have not been able to
> reliably track it down to a single cause. While trying parrallel cp -r
> (local) and grep'ing through the fs (over nfs) it eventually hung.
>
> Is there anything I can do to narrow down the source of the problem? We
> will be trying the 2.4.14 XFS 1.0.2 kernel.
>
>
> Hardware:
> 2 X GenuineIntel Pentium III (Coppermine) 996MHz
> 1Gb Memory
> Tyan Thunder HEs1 Motherboard
> 3 X 3ware Escalade 7850 IDE RAID Controller
> Intel Pro/1000 F ethernet
>
> Software:
> Redhat 7.2 w/ kernel-smp-2.4.9-13SGI_XFS_1.0.2
>
> Setup:
> Single 2TB XFS Volume, served via NFS
> No special options to mkfs.xfs.
> software RAID 0 array on top of 3 RAID 5 hardware arrays
>
>
> Thanks for any advice,
>
> -Adam
>
|