April 2022 - Mailman-users - lists.mailman3.org

Re: mail stuck in "out" queue
by Dmitry Makovey 18 Dec '24

18 Dec '24

So our mailman got stuck again, but I did follow your advise. Here's what I've found: On 3/29/19 2:58 PM, Mark Sapiro wrote: > On 3/29/19 2:28 PM, Dmitry Makovey wrote: >> I've had encountered this situation now twice: mail is being stuck in >> "out" queue and nothing in the logs indicates the issue. Is there a way >> to ask mailman-core politely to attempt delivery (perhaps with increased >> debugging in the logs) *without* restarting the core (which does solve >> *this* problem but creates other problems, like availability of the >> mailman as a whole) > > > Is the "out" runner running? yes: # docker top mailman-core | grep -F out: 100 21681 13215 0 Mar28 ? 01:14:16 /usr/local/bin/python /usr/local/bin/runner --runner=out:0:1 -C /etc/mailman.cfg > ps -fwwA |grep runner=out > > If not, you should be able to start it as the Mailman user with a > command like > > /path/to/python3 /path/to/mailman/bin/runner -C /path/to/mailman.cfg > --runner=out:0:1 > > See 'ps -fwwA|grep runner' for examples > > If the out runner is running, is there a file named *.bak along with all > the *.pck files in queue/out/ there is one, yes: # ls -l /srv/mail/mailman/core/var/queue/out/*.bak -rw-rw---- 1 100 65533 21512 Apr 14 19:45 /srv/mail/mailman/core/var/queue/out/1555296331.8386078+2c905825b82708e5a682b872b0aff320c9dd0b2e.bak what does it indicate and what should my steps be to resolve this (other than restarting queues)? > And if it's not, is there anything in Mailman's logs to indicate why it > died? > > Also, 'mailman restart' should only take a matter of seconds, but if the > out runner is running but not processing its queue, I think sending it a > SIGINT should restart it. SIGINT killed the process allright, but it didn't auto-respawn. I ended up spawning it with: /usr/local/bin/python /usr/local/bin/runner --runner=out:0:1 -C /etc/mailman.cfg but soon realized it doesn't go to background nicely so I've nohup'ed it in the end. but it did solve my problem. -- Sr System and DevOps Engineer SoM IRT

6 17

How to Increase outgoing mail volume for Mailman 3
by Brian Carpenter 09 Dec '24

09 Dec '24

Hello everyone, We had a new client come on board that brought a list that had the largest out-going mail volume that I have ever dealt with. So I did what any sane server administrator would do that is running Mailman 3: I ran to Mark Sapiro for help! Here is what the Sage of all things Mailman counseled me to do: 1. Disabled VERP delivery. VERP can help improve mail delivery by delivering 1 message per SMTP transaction (I think I got that right). Using VERP is counter-productive as it can seriously delay outgoing mail delivery for a high-volume list. So you can use something like the following in the MTA section of your mailman.cfg: [mta] verp_confirmations: no verp_personalized_deliveries: no verp_delivery_interval: 0 2. I configured Mailman 3 to run 4 outgoing runners. The default is 1. This action was the game changer. It allows Mailman to handle 4 incoming messages at a time instead of 1. My client's incoming list volume was high enough where it was overwhelming the single outgoing runner. Here is what I added to the mailman.cfg file: [runner.out] instances: 4 Doing this step immediately solved the backlog in var/queue/out. 3. If you start up with a new server for a list that is high volume then keep an eye out for Proofpoint and Vade Security as they will block your new IP address quicker than Mark Sapiro can say "This is a bug. See <some Gitlab.com url>" Mark, thank you! -- Brian Carpenter Harmonylists.com Emwd.com

5 10

Logging subscribes/unsubscribes
by Brian Carpenter 22 Oct '24

22 Oct '24

Does Mailman 3 log subscribes and unsubscribes for each list the way Mailman 2 did? Brian

5 7

Questions regarding message acceptance
by Pawel Grzywaczewski 19 Sep '24

19 Sep '24

Hello, I have few questions regarding usage of Mailman3: 1) We have list1 and list2 with members. Can we specify that all members of list1 can post messages to list2? We could potentially write a script that would synchronize members of another list to “Accept these non-members” of another list, but with thousands of lists it will be difficult to maintain lists in synchro. So perhaps there is a concept of adding another list to “Accept these non-members” and mailman real-time evaluate membership of another list upon message arrival? 2) In explanation of “Accept these non-members” it’s written: “While non-regexp addresses can be entered here, it is preferred to add the address as a nonmember and set the nonmember's Moderation to Default Processing”. I noticed that if someone posts to a list then he is added automatically to non-members. Which means that next time he will be able to post. My understanding is that it’s better to use the feature “Accept these non-members”. Am I correct? Thank you, Paweł

5 7

Display the last post time for a list or all lists.
by Henrik Rasmussen 09 Aug '24

09 Aug '24

Does Mailman 3 have something similar to https://www.msapiro.net/scripts/last_post.py displaying the last post time for a list or all lists?

3 4

MM3 generates long lines?
by peter＠chubb.wattle.id.au 28 Jul '24

28 Jul '24

I'm seeing emails being dropped when mm3 talks to the local smtp server, with log errors like: 2021-06-23 15:42:24 1lvvea-007r8u-VN H=localhost (<FQDN>) [::1] F=<announce-bounces@<FQDN>> rejected after DATA: maximum allowed line length is 998 octets, got 1036 The email itself did not have such long lines AFAICT. I can disable line length checking on the mailserver, but that sounds like a recipe for the next one along to drop the email. Peter C

4 8

mailman3 digest sending questions and also adding docker to the mix
by bob B 17 Apr '24

17 Apr '24

So I have been down the digest rabbit hole. So to confirm my assumptions/findings: Digest size threshold---- means when the total size of held messages hits this setting, a digest is sent Digest Volume Frequency. ---- specifies when the digest volume number gets incremented for digest messages, IT DOES NOT send out digests based on the setting, daily, weekly ... So what does the GUI setting "Send Digest Periodically yes no" do? how periodically does it send digests, if this is set to yes? If I want to send a digest out for a list every day, or every week, how do I set that, especially in the docker environment? Is this something that can be configured in the GUI and configured by each of my list admins? I found that ”'mailman digests --periodic' via cron” should be set in googling, but does that mean I have to do a cron job for any list that wants a scheduled digest sent? My use case is, that I have a list admin who set a size threshold, but regardless of that, they also want their list digest sent out daily, they thought it was the "Digest Volume Frequency" setting but that is not correct. However in their testing they are not getting daily digest, they are getting digests only when the threshold is reached, BUT they have "Send Digest Periodically" set to "yes" and "Enable Digests" is set to "yes"

7 17

hyperkitty.lib.utils Failed to update the fulltext index
by Ben Harris 13 Apr '24

13 Apr '24

hello there, i migrated to a new host last night and i've been getting this error from a cronjob mail every minute. anyone have any ideas what would cause this? ERROR 2019-04-17 14:26:33,056 27205 hyperkitty.lib.utils Failed to update the fulltext index: ord() expected a character, but string of length 0 found Traceback (most recent call last): File "/usr/lib/python2.7/dist-packages/hyperkitty/lib/utils.py", line 186, in run_with_lock fn(*args, **kwargs) File "/usr/lib/python2.7/dist-packages/hyperkitty/search_indexes.py", line 87, in update_index update_cmd.update_backend("hyperkitty", "default") File "/usr/lib/python2.7/dist-packages/haystack/management/commands/update_index.py", line 276, in update_backend last_max_pk=max_pk) File "/usr/lib/python2.7/dist-packages/haystack/management/commands/update_index.py", line 97, in do_update backend.update(index, current_qs, commit=commit) File "/usr/lib/python2.7/dist-packages/haystack/backends/whoosh_backend.py", line 178, in update self.setup() File "/usr/lib/python2.7/dist-packages/haystack/backends/whoosh_backend.py", line 126, in setup self.index = self.storage.open_index(schema=self.schema) File "/usr/lib/python2.7/dist-packages/whoosh/filedb/filestore.py", line 176, in open_index return indexclass(self, schema=schema, indexname=indexname) File "/usr/lib/python2.7/dist-packages/whoosh/index.py", line 421, in __init__ TOC.read(self.storage, self.indexname, schema=self._schema) File "/usr/lib/python2.7/dist-packages/whoosh/index.py", line 632, in read check_size("int", _INT_SIZE) File "/usr/lib/python2.7/dist-packages/whoosh/index.py", line 626, in check_size sz = stream.read_varint() File "/usr/lib/python2.7/dist-packages/whoosh/filedb/structfile.py", line 191, in read_varint return read_varint(self.read) File "/usr/lib/python2.7/dist-packages/whoosh/util/varints.py", line 102, in read_varint b = ord(readfn(1)) thanks, ben

5 8

Splitting hosts: mta&core vs postorius
by Ruth Ivimey-Cook 18 Jan '24

18 Jan '24

I'm trying to set up mailman3 on two hosts, one of which runs core + exim, the other runs postorius (&possibly hyperkitty). I have I think got core set up correctly, and postorius is sort of usable, but anything in the web ui that talks to core ends up with a white screen. I think what is happening is a postorius URL ends up with the core (or perhaps no) hostname, and as there's nothing there to answer it fails. I've checked the settings and I don't think I have misconfigured postorius. Is it possible that the bad url is being passed from core to postorius in some way and reused without changing the host? In the config at present, both hosts are visible to the browser, and each host can communicate with the other. Any ideas on how to pin down the problem better? Ruth

6 31

Loops back to myself
by David Krantz 12 Jan '24

12 Jan '24

Hi, I have a problem with my Mailman 3-setup. The symptom is that postfix rejects non-existing addresses for my list domains with <adr(a)example.com>: mail for example.com loops back to myself What I guess happens happens is that postfix for some reason accepts the mail as it is for a relay domain and finds that it is not deliverable locally as the address does not exist locally and then tries to relay it on to the appropriate MX server, which is itself and then fails and generates a bounce message. This is a bit annoying as I have set up postfix to only accept valid addresses. I have set the transport_maps, local_recipient_maps and relay_domains to the files generated by MM3. I run an older version of MM3 (3.0) but prefer to fix this without upgrading as upgrading is a relatively big undertaking. This is mostly a backscatter problem for spam I guess. Any ideas of very simple and easy to forget things that I've probably missed? Cheers // David

5 6