[lkml]   [1997]   [Feb]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: SCSI problems still there
    Date (Leonard N. Zubkoff) writes:
    > From: "Stephen Davies" <>
    > SCSI support in 2.0.29 has changed from earlier versions but still fails
    > for me when I try to tar a SCSI disk to a SCSI tape.

    This sounds similar to what I'm experiencing
    with a Buslogics 948 controller, a couple of SCSI
    disks and an HP Surestore tape drive (SCSI). After
    some boneheaded termination issues since resolved,
    I'm still getting unhappy results. When I do a
    backup across the net to the tape, I don't seem to
    be having any problems. However, when I do a local
    backup I've been getting intermittent lockups with
    damaged filesystems, but also so interesting
    garbage in the kernel logs, like:

    /* a whole bunch of these, 8:01 is the system disk */
    attempt to access beyond end of device
    08:01: rw=0, want=1952542573, limit=987966

    /* another boo boo with kfree, interestingly the same address as
    a previous hang, and the next one */
    kfree of non-kmalloced memory: 017cf018, next= 3a525058, order=1330793472
    kfree of non-kmalloced memory: 0247c018, next= 742d6f77, order=1718558835

    Note in both cases, it looks like ascii
    garbage. In the first case, it is
    make_request/ll_rw_blk.c that is finding the error
    in the buffer_header, and the second is
    kfree/kmalloc.c which is finding a corrupt
    page_descriptor. Since a great deal of what the
    tape is writing is ascii files, this looks very
    suspiciously like either a buffer overwrite or a
    bogus/stale pointer kind of error. (I suppose I
    could turn on SADISTIC_KMALLOC to find out if it's
    the stale variety).
    I've recompiled my kernel to dump out the first
    16 bytes of the header to see if it actually looks
    familiar when the error occurs, so we'll see.

    > If you look at the changes in 2.0.29, you will see that there have been no
    > changes to the SCSI subsystem nor to the Adaptec 1542 driver that would explain
    > this lockup problem. In fact, 2.0.29 contains very few changes of any kind.

    If this is the same problem, it would tend to
    implicate either the generic SCSI code or st.c
    since I'm getting similar unhappiness on the
    BusLogics controller. I've been send Kai mail with
    as much info as I can dig up about my
    configuration and what I'm actually seeing. If you
    know of anything else which might be suspicious,
    I'd be happy to help track it down.
    Michael Thomas (
    "I dunno, that's an awful lot of money."

     \ /
      Last update: 2005-03-22 13:39    [W:0.020 / U:14.888 seconds]
    ©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site