[Mailman-Users] Messages not arriving - deep into filtering and delivery

[Drew Tenenholz]

All --

How is the unique message ID for an outbound message created?  I'm seeing a differences in the domain label attached to outgoing messages, and I wonder if this is preventing mail from arriving to a particular (and maybe other) users.

One of my users is having trouble receiving mail from a couple of lists.  She gets mail from a couple other lists ont he same server, so we're trying to figure out what the problem is.  The current prime suspects are outbound filtering on the network where the server is located, or inbound filtering a tthe ISP (above the user's account, since nothing is going into her 'spam' mailbox).

This is a concern, since she is a principal contributor to our announce-only list (so is paying very close attention to mail sent in and mail received), and it may indicate that messages from this server (or specific lists) are being blocked, which would have an effect on other subscribers.

I've been through the mailman logs and the MTA logs, and I can see that we are, indeed sending mail to her address.  It is just not arriving.  I wonder if the following has anything to do with it.

So, in doing all of this analysis, I came accross something odd.  Two Mesages that DID ARRIVE look like this in my /var/log/maillog:

Mar 31 21:01:31 promedmail postfix/cleanup[32767]: 27A9F758035: message-id=<5158a425.6590700a.0879.ffffb894 at mx.google.com>
Mar 31 21:01:31 promedmail postfix/qmgr[4400]: 27A9F758035: from=<promed-fra-bounces at promedmail.org>, size=14135, nrcpt=311 (queue active)
...
Mar 31 21:01:32 promedmail postfix/smtp[307]: 27A9F758035: to=<pollackmp at mindspring.com>, relay=tummms.chboston.org[134.174.12.42]:25, delay=1.2, delays=0.73/0.07/0.04/0.37, dsn=2.0.0, status=sent (250 2.0.0 1ayt400cs0-1 Message accepted for delivery)

and

Apr  2 02:29:52 promedmail postfix/cleanup[6635]: 54981758035: message-id=<515A428B.50808 at gmail.com>
Apr  2 02:29:52 promedmail postfix/qmgr[4400]: 54981758035: from=<promed-eafr-bounces at promedmail.org>, size=8554, nrcpt=480 (queue active)
...
Apr  2 02:29:52 promedmail postfix/qmgr[4400]: 54981758035: from=<promed-eafr-bounces at promedmail.org>, size=8554, nrcpt=480 (queue active)


while messages that DID NOT ARRIVE all sort of look like this:

Mar 30 04:00:35 promedmail postfix/cleanup[18138]: ACCE0758034: message-id=<201303300400.r2U402Tx030401 at healthmap.chip.org>
Mar 30 04:00:35 promedmail postfix/qmgr[4400]: ACCE0758034: from=<promed-daily-update-bounces at promedmail.org>, size=9316, nrcpt=500 (queue active)
...
Mar 30 04:00:36 promedmail postfix/smtp[18143]: ACCE0758034: to=<pollackmp at mindspring.com>, relay=tummms.chboston.org[134.174.12.42]:25, conn_use=8, delay=2.1, delays=1.1/0.46/0/0.55, dsn=2.0.0, status=sent (250 2.0.0 1b42w6x590-8 Message accepted for delivery)


OK, if you've gotten this far, you'll see that the message-id ends in either mx.google.com or gmail.com (which are popular and common, and go through) and failing messages end in my specialized domain healthmap.chip.org.  Is this the sort of thing that causes a problem with delivery?  Are we maybe missing some sort of PTR records in the DNS?  That is fixable, but I don't understand why some messages get the 'special' domain and others do not.

Thanks for any help you can offer,
Drew Tenenholz