Ok, I went back and investigated my /var/log/messages.0 (the ones for last week. Here is what I found for the morning of the 26th. Previous to this there are nothing but "mark" notations for quite a while. the only thing I can trace this to is that I inadvertently started up cdcd at one point which would have involved hdd. At the time I did this, I was also doing a recording to my hard drive from a cassette recorder with the "record" from the oss package I bought from opensound.com. this is probably the reason for the comment about over-runs. May 26 01:07:18 maranatha kernel: hdd: DMA disabled May 26 01:07:28 maranatha kernel: scsi : aborting command due to timeout : pid 40675, scsi0, channel 0, id 1, lun 0 0x00 00 00 00 00 00 May 26 01:07:38 maranatha kernel: scsi : aborting command due to timeout : pid 40675, scsi0, channel 0, id 1, lun 0 0x00 00 00 00 00 00 May 26 01:07:38 maranatha kernel: SCSI host 0 abort (pid 40675) timed out - resetting May 26 01:07:38 maranatha kernel: SCSI bus is being reset for host 0 channel 0. May 26 01:07:39 maranatha kernel: scsi : aborting command due to timeout : pid 40675, scsi0, channel 0, id 1, lun 0 0x00 00 00 00 00 00 May 26 01:07:39 maranatha kernel: SCSI host 0 abort (pid 40675) timed out - resetting May 26 01:07:39 maranatha kernel: SCSI bus is being reset for host 0 channel 0. May 26 01:07:39 maranatha kernel: scsi : aborting command due to timeout : pid 40675, scsi0, channel 0, id 1, lun 0 0x00 00 00 00 00 00 May 26 01:07:39 maranatha kernel: SCSI host 0 abort (pid 40675) timed out - resetting May 26 01:07:39 maranatha kernel: SCSI bus is being reset for host 0 channel 0. May 26 01:07:40 maranatha kernel: scsi : aborting command due to timeout : pid 40675, scsi0, channel 0, id 1, lun 0 0x00 00 00 00 00 00 May 26 01:07:40 maranatha kernel: SCSI host 0 abort (pid 40675) timed out - resetting May 26 01:07:40 maranatha kernel: SCSI bus is being reset for host 0 channel 0. This goes on for a while, and then: May 26 01:09:20 maranatha kernel: ide1: reset timed-out, status=0x80 Then: May 26 02:14:35 maranatha kernel: OSS warning: Application record caused too many recording overrun errors on audio device #0. May 26 02:14:35 maranatha kernel: Look at oss/docs/README.performance for more info. And from the next boot on, my cdroms aren't heard from again on this system, except for May 26 12:34:09 maranatha kernel: ide1: BM-DMA at 0xe408-0xe40f, BIOS setti\ ngs: hdc:DMA, hdd:DMA each time I boot. The log shows clearly that they were both functioning fine up to this point. Cheryl
