FAQ
Hello,

I'm quite new to mailman so if you need more info I'll provide them - just
tell me what you need!

I have a Debian Stable server with the Debian mailman package (Version:
2.1.5-8sarge2) installed.

Mailman works great and everything could be fine if not it was restarting
every 15 minutes:

/var/log/qrunner:
May 02 20:00:02 2006 (12030) Master qrunner detected subprocess exit
(pid: 32369, sig: 9, sts: None, class: RetryRunner, slice: 1/1) [restarting]
May 02 20:00:03 2006 (12030) Master qrunner detected subprocess exit
(pid: 32462, sig: 9, sts: None, class: BounceRunner, slice: 1/1) [restarting]
May 02 20:00:03 2006 (12030) Master qrunner detected subprocess exit
(pid: 32581, sig: 9, sts: None, class: VirginRunner, slice: 1/1) [restarting]
May 02 20:00:04 2006 (12030) Master qrunner detected subprocess exit
(pid: 32753, sig: 9, sts: None, class: IncomingRunner, slice: 1/1)
[restarting]
May 02 20:00:05 2006 (12030) Master qrunner detected subprocess exit
(pid: 1377, sig: 9, sts: None, class: CommandRunner, slice: 1/1) [restarting]
May 02 20:00:11 2006 (12030) Master qrunner detected subprocess exit
(pid: 1398, sig: 9, sts: None, class: ArchRunner, slice: 1/1) [restarting]
May 02 20:00:11 2006 (12030) Master qrunner detected subprocess exit
(pid: 1439, sig: 9, sts: None, class: OutgoingRunner, slice: 1/1) [restarting]
May 02 20:00:22 2006 (20340) RetryRunner qrunner started.
May 02 20:00:24 2006 (21508) BounceRunner qrunner started.
May 02 20:00:24 2006 (23670) CommandRunner qrunner started.
May 02 20:00:26 2006 (21931) IncomingRunner qrunner started.
May 02 20:00:27 2006 (23706) OutgoingRunner qrunner started.
May 02 20:00:27 2006 (21705) VirginRunner qrunner started.
May 02 20:00:29 2006 (23682) ArchRunner qrunner started.


10 times and the max restarts is reached and nothing works. Any hints? I
searched the web and found nothing. Give me a hint what I can tune.

Thanks!

Joern

Search Discussions

  • Carl Zwanzig at May 2, 2006 at 7:54 pm

    In a flurry of recycled electrons, Joern Allmers wrote:

    Mailman works great and everything could be fine if not it was restarting
    every 15 minutes:
    Perchange, is qrunner being run every 15 minutes in a cron job? Newer mailmans
    just start qrunner once at boot time and it deals with it's own scheduling.

    z!
  • Joern Allmers at May 2, 2006 at 8:33 pm

    Am Dienstag, 2. Mai 2006 21:54 schrieben Sie:
    In a flurry of recycled electrons, Joern Allmers wrote:
    Mailman works great and everything could be fine if not it was restarting
    every 15 minutes:
    Perchange, is qrunner being run every 15 minutes in a cron job? Newer
    mailmans just start qrunner once at boot time and it deals with it's own
    scheduling.
    I don't think so. In cron.d theres no entry for it. Any idea why the qrunners
    are stopt?

    Thanks

    joern
  • Carl Zwanzig at May 2, 2006 at 10:15 pm

    In a flurry of recycled electrons, Joern Allmers wrote:
    Perchange, is qrunner being run every 15 minutes in a cron job? Newer
    mailmans just start qrunner once at boot time and it deals with it's own
    scheduling.
    I don't think so. In cron.d theres no entry for it. Any idea why the qrunners
    are stopt?
    Could be in /etc/crontab, /var/spool/cron/mailman, or some other place...
    Pre 2.1 versions ran qrunner from cron.

    In answer to the second question, you can only have one of each qrunner
    at a time, so if you start a new master, it'll kill any others that it
    sees.

    z!
  • Mark Sapiro at May 3, 2006 at 3:31 am

    Carl Zwanzig wrote:
    In a flurry of recycled electrons, Joern Allmers wrote:
    I don't think so. In cron.d theres no entry for it. Any idea why the qrunners
    are stopt?
    In answer to the second question, you can only have one of each qrunner
    at a time, so if you start a new master, it'll kill any others that it
    sees.

    It isn't that straightforward. See
    <http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq04.068.htp>.

    In any case, I don't think that's the problem here.

    Is there any more in the qrunner log than was in the original post
    (e.g., messages from the qrunners indicating a received signal)? Is
    there anything in Mailman's error log?

    It is unlikely to be any kind of processing error as that would
    normally affect at most one runner. I note that NewsRunner wasn't in
    the log in the OP. Is it unaffected, or are you not running it?

    It is most likely some external cron process or OS condition that is
    stopping the runners, but it's difficult to say without more log info
    from Mailman or the OS.

    --
    Mark Sapiro <msapiro at value.net> The highway is for gamblers,
    San Francisco Bay Area, California better use your sense - B. Dylan
  • Joern Allmers at May 3, 2006 at 8:07 am
    Hi,
    Is there any more in the qrunner log than was in the original post
    (e.g., messages from the qrunners indicating a received signal)? Is
    there anything in Mailman's error log?
    In the error log theres nothing every 15 minutes, only errors that occured
    while I did the setup. The qrunner log shows only
    May 03 09:30:01 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14115, sig: 9, sts: None, class: VirginRunner, slice: 1/1) [restarting]
    May 03 09:30:02 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14172, sig: 9, sts: None, class: OutgoingRunner, slice: 1/1)
    [restarting]
    May 03 09:30:02 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14055, sig: 9, sts: None, class: RetryRunner, slice: 1/1) [restarting]
    May 03 09:30:03 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14202, sig: 9, sts: None, class: IncomingRunner, slice: 1/1)
    [restarting]
    May 03 09:30:03 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14215, sig: 9, sts: None, class: ArchRunner, slice: 1/1) [restarting]
    May 03 09:30:04 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14229, sig: 9, sts: None, class: CommandRunner, slice: 1/1) [restarting]
    May 03 09:30:04 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14233, sig: 9, sts: None, class: BounceRunner, slice: 1/1) [restarting]
    May 03 09:30:15 2006 (1978) RetryRunner qrunner started.
    May 03 09:30:15 2006 (1694) VirginRunner qrunner started.
    May 03 09:30:17 2006 (3437) BounceRunner qrunner started.
    May 03 09:30:18 2006 (3128) IncomingRunner qrunner started.
    May 03 09:30:18 2006 (1826) OutgoingRunner qrunner started.
    May 03 09:30:19 2006 (3414) CommandRunner qrunner started.
    May 03 09:30:19 2006 (3282) ArchRunner qrunner started.

    exact every 15 Minutes.
    It is unlikely to be any kind of processing error as that would
    normally affect at most one runner. I note that NewsRunner wasn't in
    the log in the OP. Is it unaffected, or are you not running it?
    I had commented it out as I dont need it but: the only difference is that the
    newsrunner is restarted to :-(
    There is a mailman cron file but it only does checkdbs, disabled, senddigests
    etc. but none of them every 15 Minutes. I have not found any other cronjob
    that is done every 15 minutes.
    It is most likely some external cron process or OS condition that is
    stopping the runners, but it's difficult to say without more log info
    from Mailman or the OS.
    I gave you everything from mailman I have :-) In the syslog there is also
    nothing. My imap client logs in (as it does every minute) an munin is drawing
    its statistics as it does every 5 minutes.

    Do you know how to tell mailman/qrunner to be more verbose?

    Thanks

    Joern
  • Joern Allmers at May 9, 2006 at 9:20 am
    Hi,

    Sorry for the repost but I still try to tell qrunner to be more verbose (and
    of course to stop restarting every 15min) so *any* hints?

    Thanks

    Joern

    Am Mittwoch, 3. Mai 2006 10:07 schrieb Joern Allmers:
    Hi,
    Is there any more in the qrunner log than was in the original post
    (e.g., messages from the qrunners indicating a received signal)? Is
    there anything in Mailman's error log?
    In the error log theres nothing every 15 minutes, only errors that occured
    while I did the setup. The qrunner log shows only
    May 03 09:30:01 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14115, sig: 9, sts: None, class: VirginRunner, slice: 1/1)
    [restarting] May 03 09:30:02 2006 (24268) Master qrunner detected
    subprocess exit (pid: 14172, sig: 9, sts: None, class: OutgoingRunner,
    slice: 1/1) [restarting]
    May 03 09:30:02 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14055, sig: 9, sts: None, class: RetryRunner, slice: 1/1)
    [restarting] May 03 09:30:03 2006 (24268) Master qrunner detected
    subprocess exit (pid: 14202, sig: 9, sts: None, class: IncomingRunner,
    slice: 1/1) [restarting]
    May 03 09:30:03 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14215, sig: 9, sts: None, class: ArchRunner, slice: 1/1) [restarting]
    May 03 09:30:04 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14229, sig: 9, sts: None, class: CommandRunner, slice: 1/1)
    [restarting] May 03 09:30:04 2006 (24268) Master qrunner detected
    subprocess exit (pid: 14233, sig: 9, sts: None, class: BounceRunner, slice:
    1/1) [restarting] May 03 09:30:15 2006 (1978) RetryRunner qrunner started.
    May 03 09:30:15 2006 (1694) VirginRunner qrunner started.
    May 03 09:30:17 2006 (3437) BounceRunner qrunner started.
    May 03 09:30:18 2006 (3128) IncomingRunner qrunner started.
    May 03 09:30:18 2006 (1826) OutgoingRunner qrunner started.
    May 03 09:30:19 2006 (3414) CommandRunner qrunner started.
    May 03 09:30:19 2006 (3282) ArchRunner qrunner started.

    exact every 15 Minutes.
    It is unlikely to be any kind of processing error as that would
    normally affect at most one runner. I note that NewsRunner wasn't in
    the log in the OP. Is it unaffected, or are you not running it?
    I had commented it out as I dont need it but: the only difference is that
    the newsrunner is restarted to :-(
    There is a mailman cron file but it only does checkdbs, disabled,
    senddigests etc. but none of them every 15 Minutes. I have not found any
    other cronjob that is done every 15 minutes.
    It is most likely some external cron process or OS condition that is
    stopping the runners, but it's difficult to say without more log info
    from Mailman or the OS.
    I gave you everything from mailman I have :-) In the syslog there is also
    nothing. My imap client logs in (as it does every minute) an munin is
    drawing its statistics as it does every 5 minutes.

    Do you know how to tell mailman/qrunner to be more verbose?

    Thanks

    Joern
  • Mark Sapiro at May 11, 2006 at 5:30 am

    Joern Allmers wrote:

    In the error log theres nothing every 15 minutes, only errors that occured
    while I did the setup. The qrunner log shows only
    May 03 09:30:01 2006 (24268) Master qrunner detected subprocess exit
    (pid: 14115, sig: 9, sts: None, class: VirginRunner, slice: 1/1)
    [restarting]
    <snip>.
    exact every 15 Minutes.

    The qrunners are being sent SIGKILL (sig: 9) every 15 minutes. This is
    almost certainly coming from outside of Mailman. No Mailman process
    sends SIGKILL to anything.

    Do you know how to tell mailman/qrunner to be more verbose?

    There is no switch to make it log more, and it wouldn't do any good
    anyway because the runner wouldn't be able to log anything after being
    SIGKILLed anyway.

    Does your OS have a 15 minute run time limit for 'ordinary' processes?


    --
    Mark Sapiro <msapiro at value.net> The highway is for gamblers,
    San Francisco Bay Area, California better use your sense - B. Dylan

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupmailman-users @
categoriespython
postedMay 2, '06 at 6:16p
activeMay 11, '06 at 5:30a
posts8
users3
websitelist.org

People

Translate

site design / logo © 2022 Grokbase