Some questions about da0 on USB2 (recent bad behaviour)

[Andrew Reilly]

Just a follow-up with some more information:

I now doubt that the problem that I reported in the original
message is the drive dying: I've just done some read tests (cat
largefile >/dev/null) on the other USB2-attached drive (also a
Western Digital MyBook, but this one is a USB2+Firewire one with
1TB, while the other one was just USB2 with 750G.)  I'm seeing
essentially the same behaviour on that drive, too.

That is: it seems to work fine for some fraction of a minute
(doesn't seem to be longer than a minute, anyway), and then
stops completely for several minutes (processes reading or
writing sit in "D" state in ps) and then starts again, after
logging "Request completed with CAM_REQ_CMP_ERR\nRetrying
Command".

I reckon that the duty cycle of useful behaviour is is a bit
less than a third.

Any chance this is some new badness in the USB+umass stack?

Anything that I can poke or prod to make it behave better?

Any way that I can find out where it's going awry?  I don't have
kdb in my kernel, but everything not directly connected to these
USB devices seem to be behaving themselves completely.

Oh: stoppage on the two drives doesn't seem to be
chronologically correllated.

Cheers,

-- 
Andrew