
Mailing List Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [tlug] DMARC Test
- Date: Mon, 1 Jun 2026 15:55:36 +0900
- From: "Stephen J. Turnbull" <steve@??>
- Subject: Re: [tlug] DMARC Test
- References: <25348b13-93f9-422f-9818-138a37174ea4@tiuxo.com> <27161.9412.192560.293189@Stephens-MacBook-Air.local> <5ca815ee-1656-495d-bb8e-97c9efd1e8fc@codewiz.org> <27162.43825.360032.358603@Stephens-MacBook-Air.local> <de39ba63-df17-4de4-8387-4409fd911516@codewiz.org>
Bernie Innocenti writes:
> But I found a previous message, and the From had been munged to
> "bernie@??".
Interesting. That works, I guess, but most common practice is to put
the original address in a comment or the display name, and substitute
the list posting address in From.
Note: I don't seem to have gotten this one via tlug.
> I see. Mailman 3 would support ARC:
>
> https://docs.mailman3.org/projects/mailman/en/latest/src/mailman/handlers/docs/arc_sign.html
>
> ...but we're still running Mailman 2.1.29.
> And yes, I'm aware that MM3 is a complete rewrite, so it's not a
> trivial migration.
The database schema hasn't changed much for lists, so it's pretty
close to trivial, as long as you're on the same host. There are
scripts for migrating both the lists themselves and the archives (the
latter is specific to HyperKitty). It can even be zero downtime
(except archives take a long time to process through the full-text
indexers).
The main problem is that because Mailman 3 has a User model while
Mailman 2 basically treats each subscription as a separate user,
passwords don't migrate.[1] Users have to go through the one-time key
dance once for each email address they want to use.
> Is the host new enough to run the latest MM3?
I think the current release supports Python 3.9 to 3.13 (maybe 3.12).
Mailman 3 is pure Python 3 (although you also need some supporting
software like an MTA that speaks LMTP, an SQL database, and a
full-text indexer, but those shouldn't be a problem unless you're
running SunOS, AIX, or Mac OS 9 :-), so that's all you need to check.
> We can only assume that the percentage of p=reject will keep on
> growing, as the end goal is to eventually get to 100%.
I think 100% is what the Freemail Surveillance Conspiracy wants, but I
don't think that's going to happen. But it probably can continue to
grow, and that may eventually kill mailing lists.
Footnotes:
[1] Early beta versions tried, but too many users had multiple
subscriptions and Mailman kept choosing the password the user had
forgotten. A good time was had by absolutely no list owners :-)
--
GNU Mailman consultant (installation, migration, customization)
Sirius Open Source https://www.siriusopensource.com/
Software systems consulting in Europe, North America, and Japan
Home |
Main Index |
Thread Index