Re: SCSI timeouts with Advansys driver under Linux

Robert Frey (bobf@unix.advansys.com)
Wed, 4 Dec 96 17:54 PST


> I've been experiencing some SCSI timeouts on a system with an Advansys
> ABP5140 and an Adaptec 1542. The Advansys card is the one distributed with
> the HP 4020i CD-R drive.
What version of the Linux kernel are you using? (Apparently something
later than v1.3.89, based on the reset code.) What version of the
AdvanSys driver are you using? (Printed at boot time and available
from 'cat /proc/scsi/advansys/0'.) The latest is 2.0.

> The CD-R drive is connected to the Advansys card; my hard drive is
> connected to the Adaptec. After succcessfully reading data from the CD-R
> for several minutes, I get a timeout. Here's the message:
Do you mean the single 'cat' command was running for several minutes
before the timeout? So it was reading the CD-ROM sequentially.
That would appear to be the case. According to the message below the
command timed out on a read of 16KB (8 * 2KB) about 86MB (A44A * 2KB)
into the CD.

> scsi : aborting command due to timeout : pid 7890, scsi0, channel 0, id 2,
> lun 0 Read (6) 00 a4 4a 08 00
> scsi : aborting command due to timeout : pid 7890, scsi0, channel 0, id 2,
> lun 0 Read (6) 00 a4 4a 08 00
> scsi0 channel 0 : resetting for second half of retries.
> SCSI bus is being reset for host 0 channel 0.
> advansys: advansys_reset: timeout serial number changed for request 256808

I have never seen an abort request return "timeout serial number...".
The way I've written the driver I assume different serial numbers in
an abort request to mean that the command was completed after the abort
was started. The abort request is ignored and I don't do a 'done' on the
command.

> At this point, the "cat" process which was reading from the cd drive
> blocks and becomes unkillable. Any process which calls sync() blocks
> forever and becomes unkillable.
Probably because no 'done' was ever done on the aborted command.

Perhaps Leonard can shed some light on how the serial number mismatch
occurred and comment on whether my handling of it is correct.

Bob Frey
bobf@advansys.com