Kim Kimball wrote:
SAN symptoms, for those interested
I'm seeing SCSI command timeouts and UFS log timeouts (on vice
partitions using the SAN for storage) on LUNS used for vicep's on the
Hitachi USP, and was seeing them also on the 9585 until a recent
configuration change.
At first I thought this was load related, so wrote scripts to generate a
goodly load. It turns out that even with a one second sleep between
file create/write/close operations and between rm operations the SCSI
command timeouts still occur, and that it's not load but simply activity
that turns up the timeouts.
We recently purchased Hitachi UPS to virtualize LUNs from AMS-systems behind
the UPS.
(actually for dCache as a SE in the LHC-project)
The results was a lot of SCSI-errors on the client site (linux and solaris).
Hitachi send some guys to CSC to fix that issue....
I wasn't involved in that fixing part, but IIRC the point was connected to the
caching of the UPS and the "fix" involved shortening of the SCSI-command-queue
of the clients
(In your case the AFS-filesever).
I guess if you run something like iozone on one or more of the servers you'll
find
the same SCSI-errors. I really doubt it is connected to AFS.
Contact me off-line if you need more detailed info or I can then put you in
touch with the SAN-guys
here.
T/Christof
_______________________________________________
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info
