"TIMEOUT - READ_DMA48 retrying" has suddenly started appearing (RELENG_6)

[Thierry Herbelot]

Le Tuesday 27 September 2005 11:10, Pertti Kosunen a écrit :
[SNIP problems with ATA disks]
>
> I have had same problems with >UDMA100 since 5.3-RELEASE, UDMA66 works
> perfectly. Machine boots ok with higher modes, but some disk load
> triggers the timeouts. Problems just revisited after upgrade to
> 6.0-BETA5 because atacontrol mode syntax changed.
>

<AOL> Me too </AOL>

I've started to see problems with two (relatively old) 10G disks in a RAID-1 
(with gmirror). I used to have errors like :

ad8: TIMEOUT - WRITE_DMA retrying (1 retry left) LBA=9806831

The cure *so far* has been to manually reset the box with the externe "red 
button" in addition to the command-line "shutdown" or "reboot".
(I do not see these errors since I resorted to manually reset the computer at 
each boot)

The disks are connected to a crappy HTP-366 (each is master on its channel).

	TfH