I am having trouble figuring why when a message from a non-member sent to my lists does not immediately bounce back to the sender saying, "You are not a member".
Also, I have a lot of held messages in qfiles/*/* and data/*. How can I manually process those?
Thanks,
LDB
Lawrence Bowie wrote:
I am having trouble figuring why when a message from a non-member sent to my lists does not immediately bounce back to the sender saying, "You are not a member".
If that's what you want, you need to set Privacy options...->Sender filters->generic_nonmember_action) to Reject.
Also, I have a lot of held messages in qfiles/*/* and data/*. How can I manually process those?
The data/heldmsg-<list_name>-*.pck files are messages for list_name that are waiting moderator approval. Go to the list's admindb page and accept/reject/discard them as appropriate.
In which qfiles/ directory do you find messages. Other than messages currently in process, there should only be messages in qfiles/shunt/ and maybe qfiles/retry/. If there are old messages in other queues, not all your qrunners are running.
Messages in the retry queue are failed deliveries that will be periodically retried until DELIVERY_RETRY_PERIOD (default 5 days) elapses. if there are messages older than that, RetryRunner isn't running.
Messages in the shunt queue were placed there because of some error which will have been logged in the 'error' log. They can be reprocessed with bin/unshunt, but if the underlying error hasn't been addressed, they will just be shunted again.
Also bin/dumpdb and bin/show_qfiles are useful for looking at these files.
-- Mark Sapiro <msapiro@value.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
Mark Sapiro wrote:
Lawrence Bowie wrote:
I am having trouble figuring why when a message from a non-member sent to my lists does not immediately bounce back to the sender saying, "You are not a member".
If that's what you want, you need to set Privacy options...->Sender filters->generic_nonmember_action) to Reject.
Does not work? I was hoping there was something I was missing but apparently not. I do not receive the reject messag until about 4 or 6 hours later. Which is what led into the below question.....
Also, I have a lot of held messages in qfiles/*/* and data/*. How can I manually process those?
The data/heldmsg-<list_name>-*.pck files are messages for list_name that are waiting moderator approval. Go to the list's admindb page and accept/reject/discard them as appropriate.
In which qfiles/ directory do you find messages. Other than messages currently in process, there should only be messages in qfiles/shunt/ and maybe qfiles/retry/. If there are old messages in other queues, not all your qrunners are running.
Messages in the retry queue are failed deliveries that will be periodically retried until DELIVERY_RETRY_PERIOD (default 5 days) elapses. if there are messages older than that, RetryRunner isn't running.
Messages in the shunt queue were placed there because of some error which will have been logged in the 'error' log. They can be reprocessed with bin/unshunt, but if the underlying error hasn't been addressed, they will just be shunted again.
Also bin/dumpdb and bin/show_qfiles are useful for looking at these files.
Lawrence Bowie wrote:
If that's what you want, you need to set Privacy options...->Sender filters->generic_nonmember_action) to Reject.
Does not work? I was hoping there was something I was missing but apparently not. I do not receive the reject messag until about 4 or 6 hours later. Which is what led into the below question.....
So it actually does work, but you have a problem with delayed deliveries. This is probably not a Mailman issue per se, but looking into the qfiles/ entries may indicate a Mailman problem.
Otherwise, this may be a Mailman/MTA integration issue or it may be strictly in the outgoing MTA or beyond.
Does your outgoing mail go through your ISP and if so, does the ISP limit outgoing mail somehow? Note that this can be the case even if you have your own outgoing MTA. Some ISPs intercept ALL port 25 connects and route them through their own server.
-- Mark Sapiro <msapiro@value.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
Mark Sapiro wrote:
Lawrence Bowie wrote:
If that's what you want, you need to set Privacy options...->Sender filters->generic_nonmember_action) to Reject.
Does not work? I was hoping there was something I was missing but apparently not. I do not receive the reject messag until about 4 or 6 hours later. Which is what led into the below question.....
So it actually does work, but you have a problem with delayed deliveries. This is probably not a Mailman issue per se, but looking into the qfiles/ entries may indicate a Mailman problem.
Otherwise, this may be a Mailman/MTA integration issue or it may be strictly in the outgoing MTA or beyond.
Does your outgoing mail go through your ISP and if so, does the ISP limit outgoing mail somehow? Note that this can be the case even if you have your own outgoing MTA. Some ISPs intercept ALL port 25 connects and route them through their own server.
Thank you Mark... I have full control of the MTA and I am using PostFix.
I should tell you that ...
1) I do not have bounce processing enabled within Mailman
2) The mailing lists have been migrated from older versions mailman
and have been cleaned up.
3) When typing, mailq, on the server, it does not list the corresponding
held messages. These are message Mailman has in its qfiles that PostFix
does not know about.Thanks ..
LDB
Lawrence Bowie wrote:
- When typing, mailq, on the server, it does not list the corresponding held messages. These are message Mailman has in its qfiles that PostFix does not know about.
In which qfiles? If it is qfiles/retry/*, then the message has been queued for retry because the original MTA delivery attempt resulted in a transient error. The reason is logged in Mailman's smtp-failure log.
If it is some other queue, that that queue's runner is not doing its job. Have you set QRUNNER_SLEEP_TIME in mm_cfg.py to some large value?
-- Mark Sapiro <msapiro@value.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
In a flurry of recycled electrons, Mark Sapiro wrote:
Lawrence Bowie wrote:
- When typing, mailq, on the server, it does not list the corresponding held messages. These are message Mailman has in its qfiles that PostFix does not know about.
In which qfiles? If it is qfiles/retry/*, then the message has been queued for retry because the original MTA delivery attempt resulted in a transient error. The reason is logged in Mailman's smtp-failure log.
The command 'mailq' usually lists -sendmail-'s queue. Postfix uses the postqueue command, and AFAIK mailman doesn't have a command at all.
z!
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Carl Zwanzig wrote:
The command 'mailq' usually lists -sendmail-'s queue. Postfix uses the postqueue command, and AFAIK mailman doesn't have a command at all.
Postfix is very sendmail compatible and it does install a mailq command - at least on the installations I have done over the years.
Mailman has bin/show_qfiles to 'Show the contents of one or more Mailman queue files.'
Todd OpenPGP -> KeyID: 0xD654075A | URL: www.pobox.com/~tmz/pgp
Learning is something students do, NOT something done to students -- Alfie Kohn
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: When crypto is outlawed bayl bhgynjf jvyy unir cevinpl.
iG0EARECAC0FAkPoBswmGGh0dHA6Ly93d3cucG9ib3guY29tL350bXovcGdwL3Rt ei5hc2MACgkQuv+09NZUB1oWMgCeKatZuwwoeBdWMksxr0AGQnAFPKcAn06RYg7V nW4KA78Y/8uMypG6Ud/m =m3v0 -----END PGP SIGNATURE-----
Mark Sapiro wrote:
Lawrence Bowie wrote:
- When typing, mailq, on the server, it does not list the corresponding held messages. These are message Mailman has in its qfiles that PostFix does not know about.
In which qfiles? If it is qfiles/retry/*, then the message has been queued for retry because the original MTA delivery attempt resulted in a transient error. The reason is logged in Mailman's smtp-failure log.
If it is some other queue, that that queue's runner is not doing its job. Have you set QRUNNER_SLEEP_TIME in mm_cfg.py to some large value?
No, that has been modified.
LDB
Lawrence Bowie wrote:
Mark Sapiro wrote:
Lawrence Bowie wrote:
These are message Mailman has in its qfiles that PostFix does not know about.
In which qfiles? If it is qfiles/retry/*, then the message has been queued for retry because the original MTA delivery attempt resulted in a transient error. The reason is logged in Mailman's smtp-failure log.
If it is some other queue, that that queue's runner is not doing its job. Have you set QRUNNER_SLEEP_TIME in mm_cfg.py to some large value?
No, that has been modified.
So let's see if I get the situation. A nonmember send a post to a list. A reject message is generated and is queued within Mailman and takes several hours to be delivered, meanwhile it sits in some queue in mailman. Is that correct?
In which queue do you find it?
Are there any relevant messages in any of Mailman's logs, particularly 'locks', 'qrunner', 'error' and 'smtp-failure'?
-- Mark Sapiro <msapiro@value.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
Mark Sapiro wrote:
Lawrence Bowie wrote:
Mark Sapiro wrote:
Lawrence Bowie wrote:
These are message Mailman has in its qfiles that PostFix does not know about.
In which qfiles? If it is qfiles/retry/*, then the message has been queued for retry because the original MTA delivery attempt resulted in a transient error. The reason is logged in Mailman's smtp-failure log.
If it is some other queue, that that queue's runner is not doing its job. Have you set QRUNNER_SLEEP_TIME in mm_cfg.py to some large value?
No, that has been modified.
So let's see if I get the situation. A nonmember send a post to a list. A reject message is generated and is queued within Mailman and takes several hours to be delivered, meanwhile it sits in some queue in mailman. Is that correct?
Yes, that is correct.
In which queue do you find it?
the out queue
Are there any relevant messages in any of Mailman's logs, particularly 'locks', 'qrunner', 'error' and 'smtp-failure'?
Nothing good in qrunner, locks or error. smtp-failure has legitimate failures of domain not found or rejected remote delivery (due to non-existent recipient)
Lawrence Bowie wrote:
Mark Sapiro wrote:
In which queue do you find it?
the out queue
Are there any relevant messages in any of Mailman's logs, particularly 'locks', 'qrunner', 'error' and 'smtp-failure'?
Nothing good in qrunner, locks or error. smtp-failure has legitimate failures of domain not found or rejected remote delivery (due to non-existent recipient)
So are there multiple messages in the out queue? Is there a backlog? Does the 'smtp' log indicate that processing is going as fast as it can, i.e. for sequential entries in the 'smtp' log is the time between the time stamp of one entry and that of the next roughly equal to the processing time in the first entry? And, do these times seem long.
See the 'performance' related FAQ entries. These may be out of date (6.6 in particular), but 4.11 and 6.4 may be useful.
-- Mark Sapiro <msapiro@value.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
Mark Sapiro wrote:
Lawrence Bowie wrote:
Mark Sapiro wrote:
In which queue do you find it?
the out queue
Are there any relevant messages in any of Mailman's logs, particularly 'locks', 'qrunner', 'error' and 'smtp-failure'?
Nothing good in qrunner, locks or error. smtp-failure has legitimate failures of domain not found or rejected remote delivery (due to non-existent recipient)
So are there multiple messages in the out queue? Is there a backlog? Does the 'smtp' log indicate that processing is going as fast as it can, i.e. for sequential entries in the 'smtp' log is the time between the time stamp of one entry and that of the next roughly equal to the processing time in the first entry? And, do these times seem long.
See the 'performance' related FAQ entries. These may be out of date (6.6 in particular), but 4.11 and 6.4 may be useful.
Here is a recent post to a test list and it NEVER hit my mail server as reject. It never even hit the postfix mailq.
--===============0209451863==--====================> qfiles/out/1139292738.257997+63ebc154352008cb297bd321d3651ab6f813ff18.pck Subject: test is From: test-list-owner@lists.mydomain.org To: ldb@otherdomain.org MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0562359899==" Message-ID: <mailman.2.1139292738.8997.test-list@lists.mydomain.org> Date: Mon, 06 Feb 2006 22:12:18 -0800 Precedence: bulk X-BeenThere: test-list@lists.mydomain.org X-Mailman-Version: 2.1.5 List-Id: test-list.lists.mydomain.org X-List-Administrivia: yes
--===============0562359899== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit
You are not allowed to post to this mailing list, and your message has been automatically rejected. If you think that your messages are being rejected in error, contact the mailing list owner at test-list-owner@lists.mydomain.org.
--===============0562359899== Content-Type: message/rfc822 MIME-Version: 1.0
Return-Path: <ldb@otherdomain.org> X-Original-To: test-list@lists.mydomain.org Delivered-To: test-list@lists.mydomain.org Received: from localhost (localhost.localdomain [127.0.0.1]) by lists.mydomain.org (Postfix) with ESMTP id 7692A678114 for <fsg-test-list@lists.mydomain.org>; Mon, 6 Feb 2006 22:12:17 -0800 (PST) Received: from lists.mydomain.org ([127.0.0.1]) by localhost (lists.mydomain.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 09593-02 for <fsg-test-list@lists.mydomain.org>; Mon, 6 Feb 2006 22:12:17 -0800 (PST) Received: by lists.mydomain.org (Postfix, from userid 1005) id 4FD33678118; Mon, 6 Feb 2006 22:12:17 -0800 (PST) Received: from master.otherdomain.org (v757223.fromhere.com [xxx.xxx.xx.xxx]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.mydomain.org (Postfix) with ESMTP id CD97A678114 for <test-list@lists.mydomain.org>; Mon, 6 Feb 2006 22:12:15 -0800 (PST) Received: from localhost (localhost.localdomain [127.0.0.1]) by master.otherdomain.org (Postfix) with ESMTP id BD3BF46890D for <test-list@lists.mydomain.org>; Tue, 7 Feb 2006 00:12:14 -0600 (CST) Received: from master.otherdomain.org ([127.0.0.1]) by localhost (master.ldb-jab.org [209.135.157.157]) (amavisd-new, port 10024) with ESMTP id 18443-08 for <test-list@lists.mydomain.org>; Tue, 7 Feb 2006 00:12:13 -0600 (CST) Received: by master.otherdomain.org (Postfix, from userid 51) id 978E0469253; Tue, 7 Feb 2006 00:12:13 -0600 (CST) Received: from [10.3.3.7] (pool-70-105-21-105.rich.east.verizon.net [70.105.21.105]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by master.otherdomain.org (Postfix) with ESMTP id 0467946890D for <test-list@lists.mydomain.org>; Tue, 7 Feb 2006 00:12:11 -0600 (CST) Message-ID: <43E83A73.6060707@ldb-jab.org> Date: Tue, 07 Feb 2006 01:13:07 -0500 From: LDB <ldb@otherdomain.org> User-Agent: Mozilla Thunderbird 1.0.6 (X11/20050715) X-Accept-Language: en-us, en MIME-Version: 1.0 To: test-list@lists.mydomain.org Subject: test is Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: amavisd-new at ldb-jab.org X-Spam-Checker-Version: SpamAssassin 3.0.3 (2005-04-27) on lists.mydomain.org X-Spam-Level: X-Spam-Status: No, score=-2.0 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS, SPF_PASS autolearn=ham version=3.0.3 X-Virus-Scanned: by amavisd-new-20030616-p10 (Debian) at lists.mydomain.org
sd
--===============0562359899==--
Mark Sapiro wrote:
Lawrence Bowie wrote:
Mark Sapiro wrote:
In which queue do you find it?
the out queue
Are there any relevant messages in any of Mailman's logs, particularly 'locks', 'qrunner', 'error' and 'smtp-failure'?
Nothing good in qrunner, locks or error. smtp-failure has legitimate failures of domain not found or rejected remote delivery (due to non-existent recipient)
So are there multiple messages in the out queue? Is there a backlog? Does the 'smtp' log indicate that processing is going as fast as it can, i.e. for sequential entries in the 'smtp' log is the time between the time stamp of one entry and that of the next roughly equal to the processing time in the first entry? And, do these times seem long.
nothing in the mailq at all for postfix. no backlog at all. Hardly anything is running this time of night. The mail and list server are two different servers.
See the 'performance' related FAQ entries. These may be out of date (6.6 in particular), but 4.11 and 6.4 may be useful.
Mark Sapiro wrote:
Lawrence Bowie wrote:
Mark Sapiro wrote:
In which queue do you find it?
the out queue
Are there any relevant messages in any of Mailman's logs, particularly 'locks', 'qrunner', 'error' and 'smtp-failure'?
Nothing good in qrunner, locks or error. smtp-failure has legitimate failures of domain not found or rejected remote delivery (due to non-existent recipient)
So are there multiple messages in the out queue? Is there a backlog? Does the 'smtp' log indicate that processing is going as fast as it can, i.e. for sequential entries in the 'smtp' log is the time between the time stamp of one entry and that of the next roughly equal to the processing time in the first entry? And, do these times seem long.
See the 'performance' related FAQ entries. These may be out of date (6.6 in particular), but 4.11 and 6.4 may be useful.
It just flushed like you crazy!! I changed my SMTP port in Mailman from the default to 10025 (for amavisd). That was the only real change.
I hope that did it ...
LDB
participants (4)
-
cpz＠tuunq.com -
Lawrence Bowie -
Mark Sapiro -
Todd Zullinger