I believe I have narrowed it down to the unused port on the add-in
Qlogic card. Since this machine is in heavy production environment
for home dirs it's difficult to do much, and that fact that the shop
is a 24 hour shop.

We're looking at replacing all the legacy 280Rs and V480s with new Sun
Blade 6000s.

Have a Happy Holiday everyone.

Chris

In <c96021a7-b050-408f-a9e2-217fb7096719@i29g2000prf.googlegroups.com> ChrisS <chris.scarff@gmail.com> writes:

>I believe I have narrowed it down to the unused port on the add-in
>Qlogic card. Since this machine is in heavy production environment
>for home dirs it's difficult to do much, and that fact that the shop
>is a 24 hour shop.

Ours is looking like a different problem. The qlc driver and FC loop
is used only for the two internal disks. The loop only went offline
once, but there have been regular retryable SCSI errors on one of the
disks. I'm betting that one of the disks is failing. Data are
mirrored across the two disks, so I'm not as concerned now.

--
-Gary Mills- -Unix Support- -U of M Academic Computing and Networking-

Well, I fixed my problem concerning the ONLINE/OFFLINE messages. I
went back to the basics. I just pushed on the fibre cable ends (Tx/
Rv) to ensure they were seated firmly into the card receptacle.
Though it didn't click or anything ( it was in all the way), I guess
the ends just weren't perfected aligned. It is an older 1Gbit with SC
connectors (GBIC: Infineon V23826-K305-C53, Oct 2002). There must
have been some play or movement between the wire and GBIC.

Did nothing else.

"Back to the basics" . Just like the old SCSI cables, "check the
pins", right.

Chris