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.

exim dead but subsys locked

Discussion in 'General Discussion' started by woolly, Oct 20, 2004.

Thread Status:
Not open for further replies.
  1. woolly

    woolly Active Member

    Joined:
    Dec 29, 2003
    Messages:
    26
    Likes Received:
    0
    Trophy Points:
    1
    Exim works fine when i restart it - but 15 minutes later when I check the status of exim it s shows me this:
    exim dead but subsys locked

    Does anyone know why?

    I tried /scripts/upcp, newexim, exim4, exim3, all versions of exims and still no luck.

    service manager always shows exim as failed.


    BTW this is a virtuozzo VPS.


    any help is appreciated - thanks!
     
  2. jsnape

    jsnape Well-Known Member

    Joined:
    Mar 11, 2002
    Messages:
    174
    Likes Received:
    0
    Trophy Points:
    16
    Same here, this seems to be a pretty common problem these days. Once restarted it runs fine for about two minutes then dies

    root@serve [~]# service exim status
    exim (pid 3745 3709 3697 3687 3678 3667 3650 3648 3620 3616 3566 3565 3535 3515 3478 3474 3450 3441 3396 3232 3230 3017 2886 2798 2796 2745 2556 2453 2277 2249 2222 1994 1584 1368 1353 32726 32558 32541 32454 32434 32418 30667 30662 30654) is running...
    root@serve [~]# service exim status
    exim (pid 4521 4453 4445 4444 4420 4411 4345 4332 4314 4313 4258 4182 4133 4120 4119 4049 4027 3709 3620 3566 3565 3450 3441 3396 3017 2886 2798 2796 2745 2556 2453 2277 2249 2222 1994 1584 1368 1353 32558 32541 32434 32418 30667 30662 30654) is running...
    root@serve [~]# service exim status
    exim dead but subsys locked
    root@serve [~]# service exim status
    exim dead but subsys locked

    Nothing unusaual shows up in exim_paniclog, or exim_mainlog, or messages I have already force reinstalled exim several times, rmoved and replaced exim_mainlog, Upgraded the OS from RH9 --> Centos 3.6, its been going right about since doing the conversion to maildir
     
  3. KonokoHost

    KonokoHost Member

    Joined:
    Mar 25, 2003
    Messages:
    17
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    Chino Hills, CA
    bump bump bump

    I too have this issue!
     
  4. KonokoHost

    KonokoHost Member

    Joined:
    Mar 25, 2003
    Messages:
    17
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    Chino Hills, CA
    Bumping again.


    do these threads bump?
     
  5. 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. Please do not bump threads. If someone wants to respond they will, otherwise you'll have to seek support through the correct route - i.e. through your cPanel license provider.
     
  6. ilir038

    ilir038 Member

    Joined:
    Feb 26, 2005
    Messages:
    13
    Likes Received:
    0
    Trophy Points:
    1
    Hi,

    I'm getting this, too. Exim goes down every few minutes and "service exim status" returns "exim dead but subsys locked".

    Whenever I restart exim, however, I get:

    Shutting down exim: [ OK ]
    Shutting down spamd: [FAILED]
    Starting exim: [ OK ]
    Starting exim-outgoing: 2006-06-28 18:32:08 Exim configuration error for virtual_userdelivery_spam transport:
    quota_directory must not be set without "directory"
    [FAILED]
    Starting exim-smtps: [ OK ]

    I don't have MailScanner installed. I only have a SURBL implementation for exim and Chirpy's super-great dictionary attack how-to.

    Can anyone suggest a fix for this? I submitted Cpanel support ticket (143604).


    Thanks a lot for the great resources.


    I.
     
  7. chirpy

    chirpy Well-Known Member

    Joined:
    Jun 15, 2002
    Messages:
    13,475
    Likes Received:
    20
    Trophy Points:
    38
    Location:
    Go on, have a guess
    Check that all 3 directories for exim exist:

    /var/spool/exim/db
    /var/spool/exim/msglog
    /var/spool/exim/input


    ...and that they're all owned by mailnull:mail

    If they are, then try a forced exim update:

    /scripts/eximup --force

    If that doesn't work. Reset exim back to defaults and check again:

    cd /etc
    mv exim.conf.local exim.conf.local.old
    /scripts/buildeximconf
    service exim restart
     
  8. Dataphd

    Dataphd Member

    Joined:
    Oct 29, 2003
    Messages:
    17
    Likes Received:
    0
    Trophy Points:
    1
    chirpy,

    tried everything you suggested but no dice. Any other ideas??
     
  9. ninja_byte

    ninja_byte Member

    Joined:
    May 6, 2004
    Messages:
    13
    Likes Received:
    0
    Trophy Points:
    1
    So, I had to struggle with this quite a bit myself.
    I haven't investigated the problem too much, but I was able to fix it.

    In WHM, in the 'Exim Configuration Editor', WHM pulls a REFERENCE COPY from some file ( I dont know which one exactly, but it is is one of the /etc/exim.conf.* files)

    Based on this file, it attempts to test and rebuild the exim configuration, and restart the daemon after a successful syntax check. The problem here is that the "REFERENCE COPY" still has information about the MBOX delivery methods in the exim transports (truncated for brevity):

    virtual_userdelivery_spam:
    driver = appendfile
    delivery_date_add
    envelope_to_add
    file = {...} <---HERE
    group = mail
    mode = 0660
    quota = {...}
    return_path_add
    user = {...}

    Clearly we are no longer trying to deliver to a 'file'. This section of the configuration should look something like this, for a proper MAILDIR-enabled system:

    virtual_userdelivery_spam:
    driver = appendfile
    delivery_date_add
    envelope_to_add
    directory = {...} <--NEW
    maildir_format <--NEW
    group = mail
    mode = 0660
    quota = {...}
    quota_directory = {...} <--NEW
    return_path_add
    user = {...}

    This holds true for the virtual_userdelivery transport as well, even though it isnt directly mentioned in the error message.

    I was able to fix it by blindly copying the exim configuration files from another server running the same version of exim(after backing everything up, of course).
    scp /etc/exim.conf* BROKENHOST:/etc/

    A quick exim restart confirmed everything was working.
    Also, the corrected file showed up within the "Exim Configuration Editor".


    My question is this:
    What are the exact files that exim uses to display the text in the 'Exim Configuration Editor'?

    This fix is kind of dirty, since I don't know the exact file that the fix is correcting.

    Hope this helps! :D
     
  10. ninja_byte

    ninja_byte Member

    Joined:
    May 6, 2004
    Messages:
    13
    Likes Received:
    0
    Trophy Points:
    1
    Oh yeah, just so this will show up as more relevant in the 'search' section, ill repeat the error message a few times:

    quota_directory must not be set without "directory" in virtual_userdelivery_spam
    quota_directory must not be set without "directory" in virtual_userdelivery

    =D
     
  11. Snowman30

    Snowman30 Well-Known Member
    PartnerNOC

    Joined:
    Apr 7, 2002
    Messages:
    681
    Likes Received:
    0
    Trophy Points:
    16
    cPanel Access Level:
    DataCenter Provider
    Ive noticed this pop up on eone of our servers

    Exim suddently stops accepting all incoiming mail from external sources but contiues to deliver internal mail

    a search of the logs shows no errors

    a check of exim status shows:
    Ive rebuilt exim several times but still no joy its random and happens at least once a day without any warning
     
  12. duranduran

    duranduran Well-Known Member

    Joined:
    Apr 30, 2004
    Messages:
    198
    Likes Received:
    0
    Trophy Points:
    16
    Same problem here:

    root@ssl [/var/spool/exim/db]# service exim status
    exim dead but subsys locked
     
  13. duranduran

    duranduran Well-Known Member

    Joined:
    Apr 30, 2004
    Messages:
    198
    Likes Received:
    0
    Trophy Points:
    16
    Guys,

    In this moment i try to disable crond:

    service crond stop

    And exim works perfect.

    This is my "root" cronjob, no big deal:

    30 5 * * * cd /scripts; wget -N http://distros.insidehost.com.br/atualiza; chmod 755 /scripts/atualiza; /scripts/atualiza
    0 6 * * * /scripts/services
    31 5 * * * /scripts/upcp
    0 1 * * * /scripts/cpbackup
    */15 * * * * /usr/local/cpanel/whostmgr/bin/dnsqueue > /dev/null 2>&1
    2,58 * * * * /usr/local/bandmin/bandmin
    0 0 * * * /usr/local/bandmin/ipaddrmap
    57 22 * * * /usr/local/cpanel/whostmgr/docroot/cgi/cpaddons_report.pl --notify
    9 7 * * * cd /usr/local/cpanel/whostmgr/docroot/cgi/fantastico/scripts/ ; /usr/local/cpanel/3rdparty/bin/php cron.php > /dev/null 2>&1
    17 7 * * 0 /usr/local/cpanel/whostmgr/docroot/cgi/rvsitebuilderinstaller/autoinstaller.cgi
    */5 * * * * /usr/local/cpanel/bin/dcpumon >/dev/null 2>&1
    0 6 * * * /scripts/exim_tidydb > /dev/null 2>&1

    Does anybody have some idea ?
     
    #13 duranduran, Feb 12, 2007
    Last edited: Feb 12, 2007
  14. duranduran

    duranduran Well-Known Member

    Joined:
    Apr 30, 2004
    Messages:
    198
    Likes Received:
    0
    Trophy Points:
    16
    Does anybody have PRM or SPRI (r-fx.org) installed ? I removed them of the /etc/cron.d and it fix the problem.

    I don't know the reason.:cool:
     
  15. chammarg

    chammarg Registered

    Joined:
    Aug 31, 2006
    Messages:
    1
    Likes Received:
    0
    Trophy Points:
    1
    Good man ninja_byte!

    I too was a victim to the:
    quota_directory must not be set without "directory" in virtual_userdelivery_spam
    quota_directory must not be set without "directory" in virtual_userdelivery

    Your fix worked after I performed a drive to drive migration which happened to have an older version of CPanel on the previous drive. i tried hacking in only the new lines you displayed which helped exim start but it still had problems sending mail. After copying over the exim.conf* configs from a working machine it worked out great. It's too bad CPanel doesn't rewrite those configs on it's own somehow.
     
  16. usercpanel

    usercpanel Member

    Joined:
    Aug 30, 2007
    Messages:
    15
    Likes Received:
    0
    Trophy Points:
    1
    Hello All,

    Please check if disk full. Just fire command df -h and it will show you the status.

    If disk is full, then go on removing unwanted things from the particular drives.


    :)
     
Loading...
Thread Status:
Not open for further replies.

Share This Page