[Mailman-Users] Repeat messages

[Jon Carnes]

Looks like the lock files are not working... is the "sticky" bit set
properly on the directory structure?

It looks like qrunner is kicking off every minute (as it is set to do...)
and that it is processing the message in the queue as though no other
qrunner process is running.  The only reason I can think of for it to do
that, would be the lock file not working.  Ordinarily qrunner would see the
lock file on the list and then ignore those messages.

Ahhhh... Here is a long shot - Look at the message headers and see if the
messages are looping.  Could your list be a member of your list?  That would
have the same effect.  Also, someone on the list forwarding their mail to
the list would cause similar damage - those tricky end users!

Jon Carnes
----- Original Message -----
From: "Joseph McCall" <jomccall at informaxinc.com>
To: <mailman-users at python.org>
Sent: Monday, February 25, 2002 11:02 AM
Subject: [Mailman-Users] Repeat messages


> Although I know the answer will be "check your MTA" I thought I'd throw
this
> one out to the list anyway.
>
> RH 7.1 + Postfix (chroot'd) + Mailman 2.0.8
>
> Sending out a newsletter to ~12000 subscribers. One group of addresses,
all
> at the same domain, received multiple copies of the email (~20 each).
Every
> entry in maillog is identical except for the timestamp and shows
status=sent
> (250 Ok). IP's and names replaced to protected the harassed.
>
> Feb 24 23:31:03 lists postfix/smtp[13929]: 049AA27294:
> to=<name at replaced.com>, relay=smtp.replaced.com[1.2.3.4], delay=0,
> status=sent (250 Ok)
> Feb 24 23:32:04 lists postfix/smtp[14206]: C9FC2272AC:
> to=<name at replaced.com>, relay=smtp.replaced.com[1.2.3.4], delay=1,
> status=sent (250 Ok)
> Feb 24 23:33:02 lists postfix/smtp[14206]: C1A7B272AC:
> to=<name at replaced.com>, relay=smtp.replaced.com[1.2.3.4], delay=1,
> status=sent (250 Ok)
> Feb 24 23:34:02 lists postfix/smtp[14206]: 8F5C1272AC:
> to=<name at replaced.com>, relay=smtp.replaced.com[1.2.3.4], delay=0,
> status=sent (250 Ok)
> Feb 24 23:35:02 lists postfix/smtp[14206]: 35766272AC:
> to=<name at replaced.com>, relay=smtp.replaced.com[1.2.3.4], delay=0,
> status=sent (250 Ok)
> Feb 24 23:36:03 lists postfix/smtp[14206]: EFED6272AC:
> to=<name at replaced.com>, relay=smtp.replaced.com[1.2.3.4], delay=1,
> status=sent (250 Ok)
>
> It's doing this retry every minute, per qrunner cron job I suppose.
>
> Users are only subscribed once, strings on config.db show the users have
> three entries each (which everyone appears to have).
>
> In Postfix I've altered the queue lifetime and set the minimal retry time
to
> 10 hours to limit repeated sends.
>
> To top it all off, after shutting down Postfix and clearing the queues,
the
> repeats started again once Postfix was back up. Shutdown, cleared
> everything, rebooted the server, back up, same problem. Two addresses at
the
> same domain continue to get hammered. Recursive grep'ing on mailman
> directory shows only the config.db and config.db.last file have entries
for
> the user's names.
>
> Any ideas? What else can I check?
>
> Joe McCall
> UNIX Administrator
> Informax, Inc.
> http://www.informaxinc.com
>
>
>
>
> ------------------------------------------------------
> Mailman-Users maillist  -  Mailman-Users at python.org
> http://mail.python.org/mailman/listinfo/mailman-users