blammo.doh at gmail.com
Sun Apr 16 06:34:05 UTC 2006
On 4/15/06, Captain Krypto <captainkrypto at gmail.com> wrote:
> Hey all,
> I am having some issues with my backend server rebooting when put under a
> heavy I/O load to my RAID0 array... I've done some reasearch and found that
> there is a problem with the way NFS and XFS interact with each other on an
> SMP enabled kernel.... this seems like it would be a common setup for a
> myth backend system... I was wondering if anyone else has experienced this
> problem? I am running kernel version 2.6.15-1.1833_FC4smp with a striped
> hardware RAID0 array with an NFS mount to the frontend. If I boot the
> backend with the non-SMP kernel, I am able to put the backend under heavy
> disk I/O. Based on what I've googled, it seems that this problem has been
> corrected.... is anyone else seeing this?
I had the same problem as you. I replaced 100% of the hardware, kernel
versions, JFS instead of XFS, NFS v2, v3, v4, etc.
It's not just SMP.. It's XFS + NFS + Software Raid that's the unfortunate combo.
I fought this for a long time before I gave in to HW Raid. All
problems ceased the instant I removed SW Raid from the equasion.
More information about the mythtv-users