Herb Peyerl writes:
 >  >  > Have you tried "speed=1" just for grins?
 >  > 
[ panic ]
 > 
 > umm.... I think it's a safe bet that 'speed=1' didn't cause that panic.
 > 
 > I would look closely at your scsi bus. Terminators, cable length, etc. If 
 > you have any active terminators, try plugging one in and seeing if that
 > changes the world for you.

Fair enough. I took everything else (another CDROM drive and a DLT
drive) off of the bus, plugged in the shortest cable I could find, and
switched to an active terminator. No change, problem exactly as
described in my first post at speed=2, and either a panic or a timeout
followed by a bus reset at speed=1.

I happened to have a spare SS2 chassis lying around, so I transplanted
the system into that. This machine will actually burn a disk at
speed=1, so I guess there were some hardware issues with the first
SS2.

But still, I'm getting totally unreadable CDs regardless of speed:

flame# cdrecord -v dev=0,3,0 speed=1 sparccd.iso
Cdrecord 1.10 (sparc-unknown-netbsd1.5.2) Copyright (C) 1995-2001 Jvrg Schilling
TOC Type: 1 = CD-ROM
scsidev: '0,3,0'
scsibus: 0 target: 3 lun: 0
Using libscg version 'schily-0.5'
cdrecord: Warning: using inofficial libscg transport code version (frueauf-scsi-bsd.c-1.40nb1 '@(#)scsi-bsd.c   1.40 01/03/18 Copyright 1997 J. Schilling').
atapi: 0
Device type    : Removable WORM
Version        : 2
Response Format: 2
Capabilities   : 
Vendor_info    : 'YAMAHA  '
Identifikation : 'CDR102          '
Revision       : '1.00'
Device seems to be: Yamaha CDR-100.
Using driver for Yamaha CDR-100 / CDR-102 (yamaha_cdr100).
Driver flags   : SWABAUDIO
FIFO size      : 4194304 = 4096 KB
Track 01: data   68 MB        
Total size:      78 MB (07:47.22) = 35042 sectors
Lout start:      78 MB (07:49/17) = 35042 sectors
cdrecord: Warning: blockdesc secsize 2048 differs from cap secsize 0
Current Secsize: 2048
Starting to write CD/DVD at speed 1 in write mode for single session.
Last chance to quit, starting real write in 0 seconds. Operation starts.
Waiting for reader process to fill input buffer ... input buffer ready.
Starting new track at sector: 0
Track 01:  68 of  68 MB written (fifo 100%).
Track 01: Total bytes read/written: 71761920/71761920 (35040 sectors).
Writing  time:  519.451s
Fixating...
cdrecord: Input/output error. philips fixation: scsi sendcmd: retryable error
CDB:  E9 00 00 00 00 00 00 00 01 00
status: 0x0 (GOOD STATUS)
cmd finished after 159.331s timeout 480s
Fixating time:  159.396s
cdrecord: fifo had 1168 puts and 1168 gets.
cdrecord: fifo was 0 times empty and 1095 times full, min fill was 97%.
flame# mount -t cd9660 /dev/cd0a /mnt
cd0(esp0:3:0): esp0: timed out [ecb 0xf03890d0 (flags 0x1, dleft 0, stat 0)], <state 2, nexus 0xf03890d0, phase(l 12, c 100, p 3), resid 0, msg(q 0,o 0) DMA active>
Apr 12 20:18:40 flame /netbsd: cd0(esp0:3:0): esp0: timed out [ecb 0xf03890d0 (flags 0x1, dleft 0, stat 0)], <state 2, nexus 0xf03890d0, phase(l 12, c 100, p 3), resid 0, msg(q 0,o 0) DMA active>
mount_cd9660: /dev/cd0a on /mnt: Invalid argument

Anything else I should try?

Thanks,

Dan
-- 
Dan Debertin
airboss@nodewarrior.org
www.nodewarrior.org
ignorami: n: 
The art of folding problem users into representational shapes.