Georg Hager (ghager@theorie3.physik.uni-erlangen.de)
Sat, 27 Nov 1999 17:37:11 +0100
Greetings!
I'm using an Adaptec 1350 (Shuttle EPST) adapter with kernel 2.2.10
and ppSCSI V0.91, and my parallel port is set to EPP mode.
Recently I copied a large collection of files
from an attached disk (IBM DCAS-32160) when after about 500MB
the syslog showed the following errors:
-------------------------------------------------------
Nov 25 21:55:18 hilbert kernel: epst.1: Command buffer overrun, bs=32 cb=6 db=16384 bu=1 sg=0 rd=1 lp=32 pe=0 cc=59825
Nov 25 21:55:18 hilbert kernel: epst.1: Arbitration failure, bs=32 cb=0 db=16384 bu=1 sg=0 rd=1 lp=0 pe=0 cc=59826
Nov 25 21:55:18 hilbert kernel: epst.1: Arbitration failure, bs=32 cb=0 db=16384 bu=1 sg=0 rd=1 lp=0 pe=0 cc=59827
Nov 25 21:55:18 hilbert kernel: epst.1: Arbitration failure, bs=32 cb=0 db=16384 bu=1 sg=0 rd=1 lp=0 pe=0 cc=59828
Nov 25 21:55:18 hilbert kernel: epst.1: Arbitration failure, bs=32 cb=0 db=16384 bu=1 sg=0 rd=1 lp=0 pe=0 cc=59829
Nov 25 21:55:18 hilbert kernel: epst.1: Arbitration failure, bs=32 cb=0 db=16384 bu=0 sg=0 rd=1 lp=0 pe=0 cc=59830
Nov 25 21:55:22 hilbert kernel: epst.1: Bus reset
Nov 25 21:55:27 hilbert kernel: epst.1: Command buffer overrun, bs=32 cb=6 db=16 bu=1 sg=0 rd=1 lp=32 pe=0 cc=59833
A
Nov 25 21:55:27 hilbert kernel: scsidisk I/O error: dev 08:01, sector 2097094
Nov 25 21:55:27 hilbert kernel: scsidisk I/O error: dev 08:01, sector 2097126
Nov 25 21:55:27 hilbert kernel: scsidisk I/O error: dev 08:01, sector 2097690
Nov 25 21:55:27 hilbert kernel: scsidisk I/O error: dev 08:01, sector 2080808
Nov 25 21:55:27 hilbert kernel: scsidisk I/O error: dev 08:01, sector 2097094
Nov 25 21:55:27 hilbert kernel: scsidisk I/O error: dev 08:01, sector 2080808
Nov 25 21:55:27 hilbert kernel: EXT2-fs error (device sd(8,1)): ext2_read_inode: unable to read inode block - inode=260158, b
-------------------------------------------------------
After that, the mounted disk seemed to be empty, but after power cycling
the disk and reloading the epst module it worked fine again. On the second
attempt, the error appeared again, strangely enough after roghly the
same amount of data.
I'm no expert in SCSI debugging, so: Is this a problem with epst.o or
rather with my disk? The drive works fine when attached to a `normal'
fast-SCSI host adapter.
Thanks for any advice,
Georg.
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Sat 27 Nov 1999 - 11:43:53 EST