Results 1 to 10 of 10
  1. #1
    Join Date
    Feb 2002
    Posts
    1,926

    Urgent bind help

    I don't know what's up with CPanel lately (in the last week or 2 I've been having one problem after the other, mostly been able to fix them). I just rebooted the server as one of my users had been hogging 99% of the CPU for a few hours (probably a runaway script or something). Server came back up and load was back to normal, only bind didn't come back up.

    I'm running CPanel/WHM.

    I tried rebooting bind through WHM:
    ------------------------------------------------------------------------------------
    Attempting to restart bind
    rndc: connect failed: connection refused
    ------------------------------------------------------------------------------------

    Tried restarting bind via SSH, failed.

    Ran /scripts/fixndc
    ------------------------------------------------------------------------------------Found key in named.conf ..
    Found controls in named.conf ..
    named.conf has already been fixed!
    ------------------------------------------------------------------------------------

    Compared the keys in named.conf and rndc.conf and they seem to match.

    Also tried cleaning up httpd.conf, which also didn't help.

    Looked all over the CPanel forums and couldn't find any other suggestions, though I don't seem to be the first with this problem.

    Any suggestion highly appreciated.

  2. #2
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    what happens when you do /etc/rc.d/init.d/named start restart or stop via shell? Any output? anything strange to be seen in /etc/named.conf (Sorry saw your post)

    Anything in /var/log/messages ?

    How about WHM roll back feature.. make sure you back up first though before using the rollback.. to be safe.

  3. #3
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    PM me if you want me to have a look..

  4. #4
    Join Date
    Feb 2002
    Posts
    1,926
    ------------------------------------------------------------------------------------
    Oct 8 01:27:33 core named[1711]: loading configuration: failure
    Oct 8 01:27:33 core named[1711]: exiting (due to fatal error)
    Oct 8 01:27:33 core named: named startup failed
    Oct 8 01:27:39 core named: named shutdown failed
    Oct 8 01:27:56 core sshd(pam_unix)[1575]: session opened for user root by (uid=0)
    ------------------------------------------------------------------------------------

    ------------------------------------------------------------------------------------
    [email protected] [~]# /etc/rc.d/init.d/named start
    Starting named: [FAILED]
    [email protected] [~]# /etc/rc.d/init.d/named restart
    Stopping named: [FAILED]
    Starting named: [FAILED]
    [email protected] [~]# /etc/rc.d/init.d/named stop
    Stopping named: [FAILED]
    ------------------------------------------------------------------------------------

    Named.conf
    ------------------------------------------------------------------------------------
    key "rndckey" {
    algorithm hmac-md5;
    secret "SOME SECRET HERE";
    };


    controls {
    inet 127.0.0.1 allow { localhost; } keys { "rndckey"; };
    };


    zone "5413.org" {
    type master;
    file "/var/named/5413.org.db";
    };

    AND YET MORE ZONES FOLLOW
    ------------------------------------------------------------------------------------


    It does seem to be the conf file causing the problems...

  5. #5
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    Check for typos or incorrect entries in /etc/named.conf.. maybe an entry points to a .db that isn't there in /var/named/

    Check that and get back.

    Also compare against

    /etc/named.back

  6. #6
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    Just read this from /scripts/rebuildnamedconf

    #bind 9 has a nasty bug that will crash it if
    #you have two zones that are the same with diffrent cases

    Try running /scripts/rebuildnamedconf

    - Failing that, more info required from your logs would help out.

  7. #7
    Join Date
    Feb 2002
    Posts
    1,926
    Downloaded named.conf to my PC, then replaced the content with that of named.back on the server ----> bind is back up and running. I'll run it like this till I find the error on named.conf

    Thanks Dotz

  8. #8
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    NP...

    Happy to help.

    Although maybe next time :

    cd /etc
    mv named.conf named.conf.backup
    cp named.back named.conf
    /etc/rc.d/init.d/named start

    Last edited by StevenG; 10-07-2002 at 08:14 PM.

  9. #9
    Join Date
    Feb 2002
    Posts
    1,926
    Found the little culprit: domain..com in named.conf

    3 hours of downtime thanks to a dot

  10. #10
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    LOL.. those bloody DOTS huh?


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •