The Community Forums

Interact with an entire community of cPanel & WHM users!
  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Kernel Panic Error (screenshot) - what is this ???

Discussion in 'General Discussion' started by bsasninja, Feb 1, 2007.

  1. bsasninja

    bsasninja Well-Known Member

    Joined:
    Sep 2, 2004
    Messages:
    528
    Likes Received:
    0
    Trophy Points:
    16
    Last week I had a kernel panic and today happened again. The tech at the DC sent me a screeshot of the error as you can see it attached to this message.

    I see a lot of quota errors and selinux somewhere in the screenshot. Do you think that selinux was causing the problem??

    Anyways I disabled it cause i read that there are some incompatibilities with Cpanel. By the way I´m usign Centos 4.4 with Cpanel and kernel 2.6.9-42.0.3.ELsmp.

    The weirdest thing is that Im using this kernel from a long time and it started to do this errors this month. I did a memory upgrade but i doesnt have anything to do with this.

    If there is anything else i should look into tell me. I wait comments.

    Thanks in advance!
     

    Attached Files:

    #1 bsasninja, Feb 1, 2007
    Last edited: Feb 1, 2007
  2. bsasninja

    bsasninja Well-Known Member

    Joined:
    Sep 2, 2004
    Messages:
    528
    Likes Received:
    0
    Trophy Points:
    16
    i found some similar threads but no one of help, seems that there is a Kernel bug related to quotas.

    Regards
     
    #2 bsasninja, Feb 1, 2007
    Last edited: Feb 1, 2007
  3. bsasninja

    bsasninja Well-Known Member

    Joined:
    Sep 2, 2004
    Messages:
    528
    Likes Received:
    0
    Trophy Points:
    16
    I read that Centos released a new kernel two days ago 2.6.9-42.0.8.EL and 2.6.9-42.0.8.ELsmp.
    But the changelog doesnt tell anything about the error i mentioned before.

    Anyone is using it??

    Thanks
     
  4. katmai

    katmai Well-Known Member

    Joined:
    Mar 13, 2006
    Messages:
    526
    Likes Received:
    0
    Trophy Points:
    16
    Location:
    Brno, Czech Republic
    suggestion to build the latest kernel from kernel.org. the kernel default from the distro is way insecure really.
     
  5. matthewdavis

    matthewdavis Well-Known Member

    Joined:
    Jun 26, 2003
    Messages:
    90
    Likes Received:
    0
    Trophy Points:
    6
    Location:
    NC, USA
    Thats hardly the advise that should be given. What evidence do you have to support this claim?
     
  6. matthewdavis

    matthewdavis Well-Known Member

    Joined:
    Jun 26, 2003
    Messages:
    90
    Likes Received:
    0
    Trophy Points:
    6
    Location:
    NC, USA

    That's correct. The fix didn't make it into those kernels. The .0.# releases typically involve very little code changes and are security releated fixes.
     
  7. hamadino

    hamadino Registered
    PartnerNOC

    Joined:
    May 21, 2006
    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    1
    Wait for kernel 2.6.9-42.22

    This bug is fixed in kernel 2.6.9-42.22, http://bugs.centos.org/view.php?id=1601

    The most likely cases that will produce this kernel panic are either cPanel account transfer or regular accounts backup. That is due to heavy IO generated during these processes.

    I got this panic a lot during account transfers and running /scripts/fixquotas is fixing this panic temporarily.
     
  8. Un Area

    Un Area Well-Known Member

    Joined:
    Nov 16, 2006
    Messages:
    78
    Likes Received:
    0
    Trophy Points:
    6
    question

    The one you told before is for "testing" the 2.6.9-42.22 at is not recommended for production systems.
    Do you know if the latest stable kernel at kernel.org fix the issue, or if it has been fixed already in other version at kernel.org?

    Latest at kernel.org is 2.6.20.1

    Thanks!
     
  9. xerophyte

    xerophyte Well-Known Member

    Joined:
    Mar 16, 2003
    Messages:
    216
    Likes Received:
    0
    Trophy Points:
    16
    Location:
    Canada
    Katmai, i would like to know, What evidence do you have to support this claim?

    I guess its been fixed above 2.6.11, but i am not sure, but you can always compile kernel from vanilla code and see if that fix the issue.

    This url might help you too http://kb.linuxnetworkcare.com/node/9

    hope that helps
     
  10. Un Area

    Un Area Well-Known Member

    Joined:
    Nov 16, 2006
    Messages:
    78
    Likes Received:
    0
    Trophy Points:
    6
    Fixed ??

    Seems that it has been fixed in release 2.6.17 at kernel.org

    http://www.kernel.org/pub/linux/kernel/v2.6/ChangeLog-2.6.17 (below is the text)

    [PATCH] Fix oops in invalidate_dquots()

    When quota is being turned off we assumed that all the references to dquots
    were already dropped. That need not be true as inodes being deleted are
    not on superblock's inodes list and hence we need not reach it when
    removing quota references from inodes. So invalidate_dquots() has to wait
    for all the users of dquots (as quota is already marked as turned off, no
    new references can be acquired and so this is bound to happen rather
    early). When we do this, we can also remove the iprune_sem locking as it
    was protecting us against exactly the same problem when freeing inodes
    icache memory.

    Signed-off-by: Jan Kara <jack@suse.cz>
    Signed-off-by: Andrew Morton <akpm@osdl.org>
    Signed-off-by: Linus Torvalds <torvalds@osdl.org>

    Tell me if Im wrong.

    Thanks
     
  11. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    IIRC RedHat fixed the issue in 2.6.9-42.0.10.EL for those that prefer to stick with the OS vendor kernels.
     
  12. mickalo

    mickalo Well-Known Member

    Joined:
    Apr 16, 2002
    Messages:
    765
    Likes Received:
    2
    Trophy Points:
    18
    Location:
    N.W. Iowa
    are those the same kernels supplied via up2date on a RHE Linux ES release 4 O/S ??

    Mickalo
     
  13. mickalo

    mickalo Well-Known Member

    Joined:
    Apr 16, 2002
    Messages:
    765
    Likes Received:
    2
    Trophy Points:
    18
    Location:
    N.W. Iowa
    I assume it's safe then to upgrade from 2.6.9-42.0.3 to 2.6.9-42.0.10 then ... or no ??

    Thx's
    Mickalo
     
  14. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    Yes, they are. They're also used by CentOSv4 via yum.
     
  15. mickalo

    mickalo Well-Known Member

    Joined:
    Apr 16, 2002
    Messages:
    765
    Likes Received:
    2
    Trophy Points:
    18
    Location:
    N.W. Iowa
    Thx's Chirpy, always appreciate your feedback.

    Now when you say "backups" you mean the ability to reboot from a previous kernel version loaded in the grub.conf , IE: change the default=0 to default=1 to reboot to a backup kernel ?

    Mickalo
     
  16. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    Should be fine. I'd always recommend checking that you have up to date backups beforehand, just incase. It's very rare to have boot problems with a new RHE kernel, but they can happen.
     
  17. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    No, I meant "worst case scenario backups". Upgrading a kernel has the potential to FUBAR your bootloader and can render the server unbootable (without local console + recovery disk access) and so it's a good idea to have good cPanel backups.

    I'm not trying to scaremonger (too much) but it can happen on rare occasions (more frquently if you compile a kernel by hand and "forget something".

    That said, you bring up an important point in that when you use up2date or yum to upgrade a kernel, always use the install option not the upgrade option. The upgrade option may replace your existing kernel leaving you with only the new one to boot off. The install option allows you to boot back into the older kernel(s) if there's a problem booting into the new one (e.g. hardware compatibility issues).
     
  18. mickalo

    mickalo Well-Known Member

    Joined:
    Apr 16, 2002
    Messages:
    765
    Likes Received:
    2
    Trophy Points:
    18
    Location:
    N.W. Iowa
    very good information ;)

    thx's again.

    Mickalo
     
  19. bsasninja

    bsasninja Well-Known Member

    Joined:
    Sep 2, 2004
    Messages:
    528
    Likes Received:
    0
    Trophy Points:
    16
    ?¿

    sorry Chirpy but I dont understand where is the bug fixed??
    In the version 2.6.9-42.0.10 i only see this 3 bugs fixed but nothing related to the fs/quota.c bug i said when I started the thread.

    http://www.linuxsecurity.com/content/view/127223/

    223129 - CVE-2007-0001 kernel panic watching /etc/passwd
    223818 - kernel panic in sctp module
    227495 - CVE-2007-0006 Key serial number collision problem

    Tell me if Im looking at bad changelog.
    By the way Im using dual processors, so this fixes are also applied to EL.smp version too isnt it?

    Thanks
     
  20. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    I recalled incorrectly (I did say IIRC).
     
Loading...

Share This Page