Re: md0 AND md2 RAID not detected anymore

From: olli (olito24_at_gmx.de)
Date: 03/30/05

  • Next message: Iwan van der Kleyn: "Compiling vnc2swf on Debian (testing)"
    To: debian-user@lists.debian.org
    Date: Wed, 30 Mar 2005 21:02:42 +0200
    
    

    hello joey,

    yes, it seems like i'm using udev, dpkg -s udev says installed ok. but
    in
    fact i didn't even know what it was before following your link. but i
    stumbled over comments regarding udev and devfs (was it devfs?) on
    my search for the solution to this problem. but i don't know if udev was
    updated or first-time-installed with my last upgrade because i was using
    this root-lvm-raid setup for about 3 month now and aptitudes update list
    was quite long.
    i can show you an excerpt from my syslog if i don't draw on your
    patience
    too much. it shows errors regarding devfs, i don't know if udev and
    devfs
    are related, but i never noticed such a message before.
    i have a devfs dir in etc, but believing dpkg, the package is not
    installed
    on my system.

    **** syslog excerpt ****
    Mar 28 19:26:36 flyricky syslogd 1.4.1#16: restart.
    Mar 28 19:26:36 flyricky kernel: klogd 1.4.1#16, log source =
    /proc/kmsg started.
    Mar 28 19:26:36 flyricky kernel: Cannot find map file.
    Mar 28 19:26:36 flyricky kernel: No module symbols loaded - kernel
    modules not enabled.
    Mar 28 19:26:36 flyricky kernel: valid argument.<4>devfs_mk_dev: could
    not append to parent for /disc
    Mar 28 19:26:36 flyricky kernel: devfs_mk_dir: invalid
    argument.<4>devfs_mk_dev: could not append to parent for /disc
    Mar 28 19:26:36 flyricky last message repeated 115 times
    Mar 28 19:26:36 flyricky kernel: ReiserFS: dm-0: found reiserfs format
    "3.6" with standard journal
    Mar 28 19:26:36 flyricky kernel: ReiserFS: dm-0: using ordered data mode
    Mar 28 19:26:36 flyricky kernel: ReiserFS: dm-0: journal params: device
    dm-0, size 8192, journal first block 18, max trans len 1024, max batch
    900, max commit age 30, max trans age 30
    Mar 28 19:26:36 flyricky kernel: ReiserFS: dm-0: checking transaction
    log (dm-0)
    Mar 28 19:26:36 flyricky kernel: ReiserFS: dm-0: Using r5 hash to sort
    names
    Mar 28 19:26:36 flyricky kernel: ReiserFS: dm-0: switching to journaled
    data mode
    Mar 28 19:26:36 flyricky kernel: hdd: ATAPI 48X DVD-ROM drive, 512kB
    Cache, UDMA(33)
    Mar 28 19:26:36 flyricky kernel: Uniform CD-ROM driver Revision: 3.20
    Mar 28 19:26:36 flyricky kernel: Capability LSM initialized
    Mar 28 19:26:36 flyricky kernel: md: md2 stopped.
    Mar 28 19:26:36 flyricky kernel: md: md0 stopped.
    Mar 28 19:26:36 flyricky kernel: EXT2-fs: unable to read superblock
    **** / ****

    from here on, it checks the other lvm2 volumes which are reiserfs (my
    boot partition is just a raid1 ext2, not lvm, the one giving the unable
    to
    read superblock error on which fsck fails later on).

    what confuses me is, that my boot partition doesn't get started/mounted,
    my root partition does and my home partition again does not. but home is
    still accessible (it's lvm2 too).

    however, i am able to start and mount my boot raid manually when the
    system is up.

    so, can you verify that this is the udev bug you mentioned? and if it
    is,
    should i just purge the package and everything would be okay again?

    was the kernel upgraded with the same version number because of the
    udev module?

    best regards,
    oliver

    Am 30.03.2005 um 19:10 schrieb Joey Hess:

    > olli wrote:
    >> after an aptitude upgrade my md0 (/boot) and md2 (/home) raid devices
    >> are not
    >> properly detected anymore. but md1, which holds the system, is still
    >> active. besides
    >> mdadm and lvm2, the upgrade updated the kernel, but with the same
    >> version that i
    >> was already using, which is 2.6.10-1-k7. do you hav any ideas, hints,
    >> suggestions?
    >
    > Are you using udev? See
    > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=294404
    >
    > --
    > see shy jo
    >

    -- 
    To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
    with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
    

  • Next message: Iwan van der Kleyn: "Compiling vnc2swf on Debian (testing)"

    Relevant Pages

    • Re: [GIT PATCH] Remove devfs from 2.6.12-git
      ... > that udev was the official future path of the linux kernel? ... people who want dynamic /dev and persistant naming can use. ... > overridable policy in udev rather than in devfs, ...
      (Linux-Kernel)
    • devfs vs udev FAQ from the other side
      ... Presuppositions (True of both udev and devfs): ... its other names should be symlinks. ... kernel, then enabled hotplug and sysfs for udev to run. ...
      (Linux-Kernel)
    • Re: DevFS vs. udev
      ... > Just by what the udev people have said I have decided never to use it, ... Nice piece of FUD you've written without really understanding what devfs ... mostly because the support was too painful for other kernel developers. ... Device drivers do not need /dev to function, only userspace needs these. ...
      (Linux-Kernel)
    • Re: [GIT PATCH] Remove devfs from 2.6.12-git
      ... going to wait and see if udev evolved to meet my needs before devfs got ... > userspace, ... >> devfs fixes. ... say as much, not "policy in kernel". ...
      (Linux-Kernel)
    • Re: [GIT PATCH] Remove devfs from 2.6.12-git
      ... Fork the kernel? ... vocal udev proponents actively discouraged out-of-tree projects from ... Something like devfs (or the device model, ... recently heard anyone claim cryptoloop did something dm-crypt doesn't. ...
      (Linux-Kernel)