Date: Thu, 06 May 1999 21:11:00 -0700 From: Scanner <scanner@apricot.com> To: questions@freebsd.org Subject: machine crashes on heavy use of nfs exported cdrom? Message-ID: <199905070411.VAA03860@ryoohki.apricot.com>
next in thread | raw e-mail | index | archive | help
Hello,
I have been having problems where if I nfs export a cd rom
that I have mounted, and then do a fair amount of reads
on this cd the system will crash.
This is running a slightly out of date version of FreeBSD (2.2.6)
but as it is a heavily used machine and otherwise works perfectly
I have been loathe to upgrade it.
Does anyone know of a problem with FreeBSD being used
to NFS serve cd roms? If so, has this been fixed
in a later release?
Looking in /var/log/messages I found these errors
for the cd rom mounted. (This is on a Nakamichi 4 cd
scsi cdrom.. I have seen the same problem on a plain
single cd cdrom drive as well.. both scsi attached.)
May 4 11:46:27 kamidake /kernel: cd3(ahc0:2:3): SCB 0x3 - timed out while idle,
LASTPHASE == 0x1, SCSISIGI == 0x0
May 4 11:46:27 kamidake /kernel: SEQADDR = 0x4 SCSISEQ = 0x12 SSTAT0 = 0x5 SSTA
T1 = 0xa
May 4 11:46:27 kamidake /kernel: cd3(ahc0:2:3): Queueing an Abort SCB
May 4 11:46:27 kamidake /kernel: cd3(ahc0:2:3): Abort Message Sent
May 4 11:46:27 kamidake /kernel: cd3(ahc0:2:3): SCB 3 - Abort Completed.
May 4 11:46:27 kamidake /kernel: cd3(ahc0:2:3): no longer in timeout
May 4 11:46:57 kamidake /kernel: cd3(ahc0:2:3): SCB 0x3 - timed out while idle,
LASTPHASE == 0x1, SCSISIGI == 0x0
May 4 11:46:57 kamidake /kernel: SEQADDR = 0x8 SCSISEQ = 0x12 SSTAT0 = 0x5 SSTA
T1 = 0xa
May 4 11:46:57 kamidake /kernel: cd3(ahc0:2:3): Queueing an Abort SCB
May 4 11:46:57 kamidake /kernel: cd3(ahc0:2:3): Abort Message Sent
May 4 11:46:57 kamidake /kernel: cd3(ahc0:2:3): SCB 3 - Abort Completed.
May 4 11:46:57 kamidake /kernel: cd3(ahc0:2:3): no longer in timeout
May 4 11:46:57 kamidake /kernel: fhtovp: bread error 5
May 4 11:47:27 kamidake /kernel: cd3(ahc0:2:3): SCB 0x3 - timed out while idle,
LASTPHASE == 0x1, SCSISIGI == 0x0
May 4 11:47:27 kamidake /kernel: SEQADDR = 0x7 SCSISEQ = 0x12 SSTAT0 = 0x5 SSTA
T1 = 0xa
May 4 11:47:27 kamidake /kernel: cd3(ahc0:2:3): Queueing an Abort SCB
May 4 11:47:27 kamidake /kernel: cd3(ahc0:2:3): Abort Message Sent
May 4 11:47:27 kamidake /kernel: cd3(ahc0:2:3): SCB 3 - Abort Completed.
May 4 11:47:27 kamidake /kernel: cd3(ahc0:2:3): no longer in timeout
May 4 11:48:05 kamidake /kernel: cd3(ahc0:2:3): SCB 0x3 - timed out while idle,
LASTPHASE == 0x1, SCSISIGI == 0x0
May 4 11:48:05 kamidake /kernel: SEQADDR = 0x8 SCSISEQ = 0x12 SSTAT0 = 0x5 SSTA
T1 = 0xa
May 4 11:48:05 kamidake /kernel: cd3(ahc0:2:3): Queueing an Abort SCB
May 4 11:48:05 kamidake /kernel: cd3(ahc0:2:3): Abort Message Sent
May 4 11:48:05 kamidake /kernel: cd3(ahc0:2:3): SCB 3 - Abort Completed.
May 4 11:48:05 kamidake /kernel: cd3(ahc0:2:3): no longer in timeout
May 4 11:48:29 kamidake /kernel: cd3(ahc0:2:3): MEDIUM ERROR info?:0x262ae asc:
11,5 L-EC uncorrectable error
If I just use the cdrom locally and not through NFS these errors do
not show up. I did things like a "find . -exec md5 {} \;" to
stress the cd.
The scsi controller is an Adaptec 2940UW.
--Scanner (scanner@apricot.com)
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199905070411.VAA03860>