7.2-RELEASE-p4, IO errors & RAID1 failure

[Pieter de Boer]

Hi Matthew,

> I'm running 7.2-RELEASE-p4 on an i386 HP server (ML G5) in RAID1
> configuration. Very recently, I've seen IO errors such as:
>
> ad0: TIMEOUT - READ_DMA retrying (1 retry left) LBA=20472527
>
> reported and the RAID mirror is now offline.
>
> ad0: TIMEOUT - WRITE_DMA48 retrying (1 retry left) LBA=395032335
> ad0: FAILURE - WRITE_DMA48 status=51<READY,DSC,ERROR>
> error=10<NID_NOT_FOUND>  LBA=395032335
> ar0: WARNING - mirror protection lost. RAID1 array in DEGRADED mode
>
I had more or less the same timeout issues on my 8.0-RELEASE box on a 
Dell R300 with SATA disks. What I did was raise the ata timeout from 5 
seconds to 20. I did this by patching the kernel code while running, but 
I'm not sure you'd like that approach ;)

In http://www.freebsd.org/cgi/query-pr.cgi?pr=111023 a patch is 
presented that raises the timeouts by patching a few ATA kernel source 
files. This has been committed to RELENG_7 as well, so by upgrading your 
7.2-install to the latest RELENG_7 (or RELENG_8), you'll have that 
timeout fix.

Why ATA commands can take longer than 5 seconds although the disks 
appear to be fine.. wouldn't know ..

-- 
Pieter