Date: Wed, 15 May 1996 02:30:09 +0100 (BST) From: gpalmer@FreeBSD.ORG To: FreeBSD-gnats-submit@FreeBSD.ORG Subject: kern/1204: umount -f after SCSI reset -> reboot Message-ID: <199605150130.CAA00470@palmer.demon.co.uk> Resent-Message-ID: <199605150140.SAA01483@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 1204 >Category: kern >Synopsis: umount -f after SCSI reset -> reboot >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue May 14 18:40:01 PDT 1996 >Last-Modified: >Originator: Gary Palmer >Organization: None >Release: FreeBSD 2.1-STABLE i386 >Environment: May 15 02:05:18 palmer /kernel: ahc0 <Adaptec 2940 SCSI host adapter> rev 0 int a irq 11 on pci0:7 May 15 02:05:19 palmer /kernel: ahc0: aic7870 Single Channel, SCSI Id=7, 16 SCBs May 15 02:05:19 palmer /kernel: (ahc0:0:0): "CONNER CFP1060S 1.05GB 1823" type 0 fixed SCSI 2 May 15 02:05:19 palmer /kernel: sd0(ahc0:0:0): Direct-Access May 15 02:05:19 palmer /kernel: sd0(ahc0:0:0): 1013MB (2074880 512 byte sectors) May 15 02:05:19 palmer /kernel: (ahc0:1:0): "FUJITSU M1603S-512 6404" type 0 fixed SCSI 2 May 15 02:05:19 palmer /kernel: sd1(ahc0:1:0): Direct-Access May 15 02:05:19 palmer /kernel: sd1(ahc0:1:0): 519MB (1064268 512 byte sectors) May 15 02:05:20 palmer /kernel: (ahc0:6:0): "SONY CD-ROM CDU-8003A 1.9a" type 5 removable SCSI 2 May 15 02:05:20 palmer /kernel: cd0(ahc0:6:0): CD-ROM cd present.[326402 x 2048 byte records] >Description: The Conner drive sometimes wedges, and the AIC driver issues a SCSI bus reset. This also resets the CDROM drive. If the drive is mounted, you can't access it anymore after the bus reset. I tried to recover the situation with an unmount -f and got: May 15 02:05:17 palmer /kernel: panic: vref used where vget required May 15 02:05:17 palmer /kernel: May 15 02:05:17 palmer /kernel: syncing disks... 2 2 done May 15 02:05:17 palmer /kernel: (sorry, nothing else that is relevant). A dump was started, but I must have hit a key on the keyboard (I was in X and didn't know that it had panic'd) and the dump aborted ... is that REALLY necessary? The reset messages are: May 14 07:33:17 palmer /kernel: sd0(ahc0:0:0): timed out in dataout phase, SCSISIGI == 0x0 May 14 07:33:17 palmer /kernel: ahc0: Issued Channel A Bus Reset #2. 2 SCBs aborted May 14 07:33:17 palmer /kernel: sd0(ahc0:0:0): UNIT ATTENTION asc:29,0 May 14 07:33:17 palmer /kernel: sd0(ahc0:0:0): Power on, reset, or bus device reset occurred field replaceable unit: 14 May 14 07:33:17 palmer /kernel: , retries:3 May 14 07:33:17 palmer /kernel: sd1(ahc0:1:0): UNIT ATTENTION asc:29,0 May 14 07:33:17 palmer /kernel: sd1(ahc0:1:0): Power on, reset, or bus device reset occurred May 14 07:33:17 palmer /kernel: , retries:4 May 15 02:00:05 palmer /kernel: cd0(ahc0:6:0): UNIT ATTENTION asc:29,0 May 15 02:00:05 palmer /kernel: cd0(ahc0:6:0): Power on, reset, or bus device reset occurred I guess that /etc/daily tried to access the CDROM drive, which caused the UNIT ATTENTION, which came up on my console, and that's when I noticed it was still mounted... >How-To-Repeat: Have a CDROM mounted when an AIC controller resets the SCSI bus, and then try unmounting it. >Fix: Not my area, sorry. >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199605150130.CAA00470>