Re: disk recovery help

From: Charles Sprickman (spork_at_fasttrackmonkey.com)
Date: 07/20/04

  • Next message: Joseph M Link: "xpt_schedule() and start interface context question." 
    Date: Tue, 20 Jul 2004 14:01:06 -0400 (EDT)
To: Peter Jeremy <PeterJeremy@optushome.com.au>

    On Tue, 20 Jul 2004, Peter Jeremy wrote:

    > It's difficult to see how a sanely written RAID utility could totally
    > screw up an array in a short time - a 'build' utility presumably
    > writes known data to the array but logically it would do so
    > sequentially. The only thing I can think of is that the controller
    > had a significant amount of cached data - which has been wiped, rather
    > than written back to the disk.

    This is one case where I'd like to see a "Are you sure (y/n)?" prompt. In
    the future, I'm sticking with the adaptec BIOS to do a drive
    replace/rebuild. Adaptec's raidutil is just a bit too easy to shoot
    yourself in the foot with. I have a call in to adaptec about this, they
    are having a hard time telling me just what the "build" (not "rebuild")
    command does, but they are fairly certain that it writes it's config at
    the end of the disk, then zeros it from the outside in.

    > If you haven't run newfs and have the correct disklabel, the disk
    > should be in a reasonably sane state. Have you tried running something
    > like ports/sysutils/scan_ffs over the disk (or your copy of it)?

    I don't have the actual disk; we're very short on any sort of spare
    hardware so we had to newfs the partition in question and start over. I
    grabbed the dd "image" before that. An fsck on the problem partition ran
    for 12 hours and I don't know how far along it was. I looked at scan_ffs
    just now, and it looks like it works on the whole disk trying to find the
    label. Since I only have one partition, there's no label.

    > Have you tried dumping vn0c?

    I could try that, but I'd have to find myself another 26GB of space
    somewhere.

    Thanks!

    Charles

    > --
    > Peter Jeremy
    >
    _______________________________________________
    freebsd-hackers@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
    To unsubscribe, send any mail to "freebsd-hackers-unsubscribe@freebsd.org"

  • Next message: Joseph M Link: "xpt_schedule() and start interface context question."

    Relevant Pages

    • Re: Problem in Solaris 10 Installation - UPDATE
      ... I tried to resize backup partition so that it is: ... Ready to label disk, continue? ... In this way, I can label the disk without error, but the installation gives the same error. ...
      (comp.sys.sun.admin)
    • Re: format OK, but mount still fail.
      ... AVAILABLE DISK SELECTIONS: ... # format /dev/rdsk/c1t5d0s0 ... partition - select a partition table ... label - write label to the disk ...
      (comp.unix.solaris)
    • Re: Problem in Solaris 10 Installation - UPDATE
      ... Please delete the partition and install the OS. ... Ready to label disk, continue? ... In this way, I can label the disk without error, but the installation gives the same error. ...
      (comp.sys.sun.admin)
    • Re: newfs command failed.
      ... Then I use the following procedure to partition the new disk ... AVAILABLE DISK SELECTIONS: ... format - format and analyze the disk ... label - write label to the disk ...
      (comp.unix.solaris)
    • How to remove a label?
      ... I'm trying to remove a label from a disk so I can relabel it - the ... The disk has been put into a Sun which has a fully working boot disk, ... when I looked at the partition table, ... Partition 1 overlaps partition 2. ...
      (comp.sys.sun.hardware)