From owner-freebsd-hackers Sat Apr 22 04:46:18 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA20895 for hackers-outgoing; Sat, 22 Apr 1995 04:46:18 -0700 Received: from hda.com (hda.com [199.232.40.182]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id EAA20885 for ; Sat, 22 Apr 1995 04:46:11 -0700 Received: (dufault@localhost) by hda.com (8.6.9/8.3) id HAA03810; Sat, 22 Apr 1995 07:45:05 -0400 From: Peter Dufault Message-Id: <199504221145.HAA03810@hda.com> Subject: Reprobing scsi and bringing devices on line To: wilko@yedi.iaf.nl (Wilko Bulte) Date: Sat, 22 Apr 1995 07:45:05 -0400 (EDT) Cc: hackers@freefall.cdrom.com In-Reply-To: <199504211633.SAA01248@yedi.iaf.nl> from "Wilko Bulte" at Apr 21, 95 06:33:30 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 1848 Sender: hackers-owner@FreeBSD.org Precedence: bulk Wilko Bulte writes: > > > I've also seen 'adapter frozen' or something similar on AH154x cards. > Is this what you mean? I'm not sure, since what I usually see are lots of "sd?: timeout" messages followed by a hang since I'm running X. > > > will redo a transfer that was in progress when a device powered up on > > the bus, assuming the disk drive properly detects the power up. > > That's the way it should work yes. Don't count on all hostadapters > to keep their act together though ;-) It could be the host adapter getting screwed up, you're right. > > > A utility that suspends all activity on the SCSI bus will make powering > > up a device safe, though, and I've been thinking of doing it (suspend > > activity, beep when all activity is suspended, delay for 10 seconds, and > > permit activity to resume). > > Should work fine. The raidsystems I use at work (==DEC) use a similar > approach to 'hotswap' drives. This is working OK. I've had it running all night suspending I/O to the scsi bus for 20 seconds and then permitting it to run for 20 seconds while to loops of "cp -r /etc /usr/foo; rm -rf /usr/foo" are running > > While we are at it: the 'scsi' tool reputedly should allow a re-probe > of all devices. What is the exact syntax for this? The man page > is less than obvious to me (this is _still_ 1.1.5, I need stability > more than features (...)) There is no reprobe of all devices in 1.1.5. In -current there is. A reprobe of 1 device in 1.1.5 is effected by: scsi -f device -r [-b bus] [-t targ] [-l lun] # To reprobe a device where I think -f can be any of the "control" devices such as (for 1.1.5) /dev/rsd?d or /dev/st0ctl.0. -- Peter Dufault Real Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267