[4285] in linux-scsi channel archive
more DPT trouble
daemon@ATHENA.MIT.EDU (Jon Lewis)
Sat Jul 4 02:58:11 1998
Date: Sat, 4 Jul 1998 02:56:41 -0400 (EDT)
From: Jon Lewis <jlewis@inorganic5.fdt.net>
To: Linux SCSI Mailing List <linux-scsi@vger.rutgers.edu>
I setup remote syslog to catch the DPT error messages and started moving
big files from /mnt/sdb1 to /mnt/sdb2. It took a bit longer this time,
but eventually the machine started showing lots of errors and rebooted.
I'm using RedHat's 5.1 distribution kernel (2.0.34-0.6). Here are
the messages from loading eata_dma:
EATA (Extended Attachment) driver version: 2.59b
developed in co-operation with DPT
(c) 1993-96 Michael Neuffer, mike@i-Connect.Net
Registered HBAs:
HBA no. Boardtype Revis EATA Bus BaseIO IRQ DMA Ch ID Pr QS S/G IS
scsi1 : PM3334UW v07M.0 2.0c PCI 0xd010 11 BMST 2 7 N 64 252 Y
scsi1 : EATA (Extended Attachment) HBA driver
scsi : 2 hosts.
Vendor: DPT Model: LINUX Rev: 07M0
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sdb at scsi1, channel 0, id 0, lun 0
scsi1: queue depth for target 0 on channel 0 set to 64
SCSI device sdb: hdwr sector= 512 bytes. Sectors= 26773440 [13072 MB] [13.1 GB]
sdb: sdb1 sdb2
The SCSI bus consists of 4 IBM DDRS-34560 drives. We set it (and left it)
down to 5mhz sync when we started having trouble...but Granite assured us
the cabling we bought would be suitable for 20mhz. Termination is the
card's built-in at its end and a Granite active terminator at the other
end. We've also tried without the Granite terminator with the last drive
terminated.
Here are typical crash messages, which get quickly followed by a
spontaneous reboot.
Jul 5 02:34:23 io5 kernel: scsi1 channel 0 : resetting for second half of retries.
Jul 5 02:34:23 io5 kernel: SCSI bus is being reset for host 1 channel 0.
Jul 5 02:34:23 io5 kernel: eata_reset called pid:85145 target: 0 lun: 0 reason 0
Jul 5 02:34:23 io5 kernel: eata_reset: board reset done, enabling interrupts.
Jul 5 02:34:23 io5 kernel: eata_reset: interrupts disabled again.
Jul 5 02:34:23 io5 kernel: eata_reset: exit, wakeup.
Jul 5 02:34:23 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:23 io5 kernel: scsidisk I/O error: dev 08:12, sector 669434, absolute sector 14055162
Jul 5 02:34:23 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:23 io5 kernel: scsidisk I/O error: dev 08:12, sector 669436, absolute sector 14055164
Jul 5 02:34:23 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:23 io5 kernel: scsidisk I/O error: dev 08:12, sector 669438, absolute sector 14055166
Jul 5 02:34:23 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:23 io5 kernel: scsidisk I/O error: dev 08:12, sector 669440, absolute sector 14055168
Jul 5 02:34:23 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:23 io5 kernel: scsidisk I/O error: dev 08:12, sector 669442, absolute sector 14055170
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669444, absolute sector 14055172
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669446, absolute sector 14055174
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669448, absolute sector 14055176
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669450, absolute sector 14055178
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669452, absolute sector 14055180
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669454, absolute sector 14055182
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 671800, absolute sector 14057528
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 2, absolute sector 13385730
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 6, absolute sector 13385734
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669456, absolute sector 14055184
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669458, absolute sector 14055186
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 672560, absolute sector 14058288
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669460, absolute sector 14055188
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
Jul 5 02:34:24 io5 kernel: scsidisk I/O error: dev 08:12, sector 669462, absolute sector 14055190
Jul 5 02:34:24 io5 kernel: SCSI disk error : host 1 channel 0 id 0 lun 0 return code = 27070000
This goes on for about 10s before it reboots.
The only other thing I can think of trying is to make sure write caching
on each drive is disabled. I'll probably try that (and rebuild the raid
group) tomorrow. If that doesn't help, I may test each drive on an
NCR-875 and make sure the drives aren't bad.
------------------------------------------------------------------
Jon Lewis <jlewis@fdt.net> | Spammers will be winnuked or
Network Administrator | drawn and quartered...whichever
Florida Digital Turnpike | is more convenient.
______http://inorganic5.fdt.net/~jlewis/pgp for PGP public key____
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
