Results 1 to 11 of 11

Thread: Problem: start-stop-daemon: group 'Debian-exim' not found

  1. #1
    Join Date
    Jan 2010
    Location
    EU / PL
    Posts
    121

    Problem: start-stop-daemon: group 'Debian-exim' not found

    I got report from cron on mail:

    Topic:
    Cron <root@server> test -x /usr/sbin/anacron || ( cd / && run-parts --report /etc/cron.daily )

    Body:
    /etc/cron.daily/exim:
    start-stop-daemon: group 'Debian-exim' not found
    start-stop-daemon: group 'Debian-exim' not found No passwd entry for user 'Debian-exim'
    run-parts: /etc/cron.daily/exim exited with return code 1

    Debian 9x64
    Any suggestion how to resolve this?

    Edit:
    Part of cron which make error:

    Code:
    # run tidydb as Debian-exim:Debian-exim.
    if [ -x /usr/sbin/exim_tidydb ]; then
      cd $SPOOLDIR/db || exit 1
      if ! find $SPOOLDIR/db -maxdepth 1 -name '*.lockfile' -or -name 'log.*' \
        -or -type f -printf '%f\0' | \
          xargs -0r -n 1 \
          start-stop-daemon --start --exec /usr/sbin/exim_tidydb \
          --chuid Debian-exim:Debian-exim -- $SPOOLDIR > /dev/null; then
        # if we reach this, invoking exim_tidydb from start-stop-daemon has
        # failed, most probably because of libpam-tmpdir being in use
        # (see #373786 and #376165)
        find $SPOOLDIR/db -maxdepth 1 -name '*.lockfile' -or -name 'log.*' \
        -or -type f -printf '%f\0' | \
        su - --shell /bin/bash \
             --command "xargs -0r -n 1 /usr/sbin/exim_tidydb $SPOOLDIR > /dev/null" \
             Debian-exim
      fi
    fi
    Is this necessary? Or should I change Debian-eximebian-exim to mail:mail?
    Last edited by Magician; 11-02-2017 at 12:14 PM.
    VPS KVM + Debian 9x64 + DA
    Developer/linux user notes

  2. #2
    Join Date
    Apr 2005
    Location
    GMT +7.00
    Posts
    11,404
    Hello,

    Probably not many of us here did try Directadmin with Debian 9 server yet. So you might want to open a ticket with Directadmin support directly.
    With regards, Alex.

    Professional Server Management for web hosting companies and individuals
    Hourly Support, Disaster Recovery, Server Hardening, Monthly Subscription
    Directadmin installation and optimization

    Click here if you need a Linux Admin

  3. #3
    Join Date
    Oct 2007
    Location
    France
    Posts
    23
    On top of that, the script /etc/cron.daily/exim refers to debian4 which is the executable when exim is installed with debian packets.
    With the installation from DirectAdmin, the executable is exim, not exim4

    On line 78, it calls "exim4 root". If this is replaced by "exim root" it won't be delivered and create an entry in the paniclog as root is on the non-delivery list by default.

    On line 82, the file/etc/logrotate.d/exim4-paniclog doesn't exist.

    This script is for the debian's exim package, not for Directadmin's exim !

    I have now 9 servers running Debian 9 with directadmin. Debian9+Directadmin is as stable as debian 8 and has just a few minor bugs like this one

  4. #4
    Join Date
    Feb 2005
    Location
    The Netherlands
    Posts
    304
    Also have the same problems over here with that daily:

    Code:
    run-parts: /etc/cron.daily/exim exited with return code 123
    message, did you already got a reply back from DirectAdmin about this?

  5. #5
    I'm not entirely sure that's our file... it might be from another exim install, perhaps from apt-get.

    In any case, yes, swapping to mail:mail might do the trick... or just remove the file.
    So removing the file is probably going to be the more reliable solution.

    If it's always present on all DA installs, let us know and we can check into it, probably just automatically deleting the file after the install is done.

    John

  6. #6
    Correction, I am seeing it present in our da_exim.deb install pack for Debian 9.
    I've removed it, re-uploaded a new da_exim.deb file, and re-packed the services_debian90_64.tar.gz on files1.

    John

  7. #7
    Join Date
    Feb 2005
    Location
    The Netherlands
    Posts
    304
    Great, so we can just go ahead and remove the /etc/cron.daily/exim file?

  8. #8
    Yes, it's not on any CentOS box, so likely not a critical cron.

  9. #9
    Join Date
    Feb 2005
    Location
    The Netherlands
    Posts
    304
    So how would paniclog notifications work without this?

  10. #10
    As per all current CentOS boxes, if your email system isn't working, you'd check the logs to see why.
    This just prevents the active notice, but most people know right away when something is broken.

    I'd almost be willing to have the dataskq do a check on the file, and notify all admins, as I believe the above script would end up going to "root".
    I'm not sure how many admins actually check that.. or redirect it to admin or some other value in /etc/aliases.

    John

  11. #11
    Join Date
    Feb 2005
    Location
    The Netherlands
    Posts
    304
    This is a really non-professional way of managing servers IMHO if you administer more than one server. The active notice about a problem on one of our servers helps us to tackle problems even before a client notices.

    Wouldn't it be simpler to just fix the error or use the cron supplied by Debian?

    Every somewhat decent admin should check emails sent to root of course!

Posting Permissions

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