>Coincidence or related bugs? I've upgraded to 2.0.10 and have not seen
>it again...but it took nearly a month (I think) to show up in 2.0.3 and
>I've only had 2.0.10 up 2 weeks.
Bug methinks.
I'm getting the same swap errors and was having trouble with scsi errors
on 2.0.13 after 26-36 hours uptime, which I don't think was coincidental
as the step from 2.0.12->13 had a lot of Adaptec aic7xxx changes.
Unfortunately the process running on the console would wipe out the error
messages before I could record them and the machine would lock solid
a few minutes later - no oops, consle switching worked, as did the debug
keys, but each shell was locked totally and all incoming connections were
refused.
At the suggestion of one of the aic7xxx hackers I enabled tagged queuing
and bumped SCBs up to 8 when I stepped to 2.0.14. Unfortunately I don't
have enough uptime to say whether the problem's still here.
The box is a news server with swap on the scsi bus too, so it's generally
hammered hard, which means that any bugs should surface again fairly quickly.
Hopefully I'll be around to record them
AB
-- (3) With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. It is hard to be sure where they are going to land, and it could be dangerous sitting under them as they fly overhead. -- RFC 1925 - Fundamental Truths of Networking