[CentOS] hardware issues under 5.6

Mon May 9 08:58:01 UTC 2011
Laurent CREPET <lct at megrapet.fr>

On Sun, 08 May 2011 19:30:10 -0400
mark <m.roth at 5-cent.us> wrote:

> Folks,
> 
>     This is, umm, odd. Two, at least, and possibly a third over the 
> weekend, of some older Dell 1950's, suddenly died shortly after being 
> upgraded to 5.6. They're all about the same age, and it's not happening 
> with other systems, but they claim a scsi abort, and the f/s goes 
> read-only. The little lcd screen goes orange, and shows a fatal error E 
> 171, and B0 F3 D0? D0 F3 B0? I forget, and won't be able to double check 
> my memory till the morning.
> 
>     Has anyone seen this, with a f/s suddenly going r/o, on machines 
> that seemed to be running fine for years? Since this is two? three 
> machines, that's somewhere between "coincidence" and "enemy action".

Probably a coincidence. Last week, I lost 2 disks in the same system in 2 days. The 2nd time, even if RAID 5 status was optimal, it did not handle the loss of one hard-disk.

We have a bunch of PowerEdge 1950/2950 running CentOS 5.6. No specific problem since the upgrade (I though about an issue with Squid, but no link with the fact that I upgraded to 5.6).

If you have the LCD screen orange, this is an hardware-related issue. Some of them can be prevented/fixed by upgrading firmwares. You can easily upgrade from CentOS using Dell Server Update Utility or Dell Linux Hardware repository.

Laurent.

-- 
Laurent CREPET <lct at megrapet.fr>