Messages in this thread | | | Date | Thu, 01 Jan 1998 00:12:54 -0600 (CST) | From | Doug Ledford <> | Subject | RE: kernel: scsi : aborting command due to timeout |
| |
On 30-Dec-97 Jeffrey Hundstad wrote: >-----BEGIN PGP SIGNED MESSAGE----- > >These are logs from a couple of machines I admin at the university. >Both of these boxes are scsi machines. Doombox has (1) scsi disk >Nitrogen has (3) scsi devices where these devices are MD'ed together. >I was not interactive during either of these resets... but it didn't >seem to cause any damage. I was curious whether or not I should be >concerned.
Either you use 2.0.33 or you have a patched aic7xxx driver. In either case, the patch is mine and the intent of the patch is to do exactly what you saw: namely, survive a SCSI bus reset. The old verison of the driver would crash when it came to reset time, so if your machine had a SCSI bus reset and didn't crash then my patch did its job. Nothing more really, there are occasions when there will be reset simply because the sequencer in that version of the driver isn't perfect and some drives will occasionally hang on it (usually because it's very fast at sending commands).
---------------------------------- E-Mail: Doug Ledford <dledford@dialnet.net> Date: 01-Jan-98 Time: 00:12:55 ----------------------------------
| |