2001-10-31 17:48:32

by Evan Harris

[permalink] [raw]
Subject: Question about 2.4.x old aic7xxx scsi driver and command timeouts. Corruption?


I added an Adaptec 2940U2W adapter and several fast scsi drives to a linux
system and upgraded to the latest stock 2.4.13 kernel. I tried using the
newer aic7xxx driver, but since my bus occasionally has command timeouts, I
went back to the older driver, as the newer one handles them less
gracefully.

But, I occasionally get syslog messages like:

scsi : aborting command due to timeout : pid 2811297, scsi0, channel 0, id
4, lun 0 Write (10) 00 01 82 7d 3a 00 02 00 00
scsi : aborting command due to timeout : pid 2811302, scsi0, channel 0, id
4, lun 0 Write (10) 00 01 82 7f 3a 00 02 00 00
scsi : aborting command due to timeout : pid 2811303, scsi0, channel 0, id
12, lun 0 Write (10) 00 02 9e 16 ef 00 00 20 00
scsi : aborting command due to timeout : pid 2811304, scsi0, channel 0, id
0, lun 0 Read (10) 00 01 8c 38 88 00 01 00 00

Other than the syslog and console message everything seems to be fine other
than a few second pause when it happens.

My question is, in the old driver, when this happens, should I be worried
about data loss or corruption, or does the scsi-layer automatically reissue
the timed out command?

Evan