From: Franz Fischer (Franz.Fischer@franz-fischer.de)
Date: Tue Mar 15 2005 - 14:53:36 EST
Hi all,
thanks to Tom L, Benjamin, Selden and JEAN-PHILIPPE for their replies.
The most valuable hint came from Selden, who pointed out that ``errors on a
disk itself usually are reported as "media errors".''
This triggered me to more carefully read the uerf output:
...
SENSE KEY x000B ABORTED CMD
...
ADDITION LEN x18
CMD SPECIFIC 3 x00
CMD SPECIFIC 2 x00
CMD SPECIFIC 1 x00
CMD SPECIFIC 0 x00
ASC x1B
ASQ x00
...
and the disk drives manual, which lists ``Synchronous Data Transfer error:
extra pulses on sync transfer'' for the x000B/x1B combination. Taking this
as a further hint, I disabled target initiated synchronous negotiation via
the corresponding jumper and now diskx runs for almost one hour w/o any
problems. (Before the error frequence was in the range of minutes.)
Thus, neither the drive is (almost) dead, nor the problem has been a
``real'' cabling problem. It just seems that the DGHS18Y's target initiated
sync negotiation is somewhat to freaky for the elderly AS255.
--- original question ---
> Hi all,
>
> I was trying to move a COMPAQ DGHS18Y 18GB drive from a retiered file server
> to my Alpha box, but I see repeated hard errors reported by uerf, sometimes
> the system gets hung for a while due to SCSI bus timeouts / resets.
>
> Does this indicate the drive is (almost) dead?
>
> Current setup is AlphaStation 255, Tru64 UNIX 4.0G, narrow internal SCSI
> bus, DGHS18Y connected via 50pin to 80pin SCA adapter.
>
> [...]
Regards
\franz
-- Franz G. Fischer -------- Franz dot Fischer at franz-fischer dot de
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:50:16 EDT