I have been instructed by Mark Sapiro to subscribe to this list to post questions related to mailman3.
On Mon, 31 Jan 2022, Mark Sapiro wrote:
On 1/31/22 06:20, Lucio Chiappetti wrote:
One of the lists of which I am a co-moderator has been moved from mailman 2 to 3.
Actually the migration has been made operative yesterday (and so we got the first problems ...).
Basic background. There are a number of lists within my institution (a nationwide research institution with many seats across our country), some are institutional (all staff subscribed ex officio) and managed by staff in one seat, some other are non-institutional (subscription is optional) and are managed by a few volunteers in other cities different from the seat where the mailman server runs.
In particular, the list settings for *_these_nonmembers are at Settings -> Message Acceptance, but these settings in particular are deprecated in Mailman 3 except for regexps. In Mailman 3 the preferred way to say accept a nonmember is to create a nonmember for that user and set it's Moderation action to Default Processing. If, a nonmember posts, the nonmember record is automatically created,
Let me explain our case, and how it was managed in mailman2.
originally our staff (of the order of 1000) had addresses like user@place.inaf.it or user@place.astro.it. Some had even two (say place1.inaf.it and place2.inaf.it), so they subscribed with ONE of their work addresses
later we got addresses like name.surname@inaf.it. New users subscribe with this address but MANY of the old ones continue to use the PREVIOUS address.
we implemented a spam filtering policy as follows:
staff (and former staff) can SUBSCRIBE with any address or CHANGE their address. Any subscribed address will be recognised. Mail is NOT held in moderation unless too bulky, or has undisclosed recipients
mail from non members with any address in the .it TLD or in the tng.iac.es domain (where some of our staff work) goes into moderation (this is achieved via regexp in hold_these_nonmembers)
anything else is silently discarded as spam (generic_nonmember_action = discard)
when some staff posts from a non-subscribed address (be it the home address, or one of the alternate addresses described above) it therefore goes into moderation because of hold_these_nonmembers
in mailman2 we could let them pass "manually" (if we recognised the person) or (in case they used an alternate address) tick to add them permanently to accept_these_nonmembers
(in one tick)
at the end we got somehow fed up of this but we did NOT want to accept any address of the form ^.+@.+\.inaf.it (which could be forged) but found that we could simply put in accept_these_nonmembers two entries like @list1 @list2
where list1 and list2 are two institutional lists on the same server which contain all permanent and temporary staff or assimilated with their official address (these lists are automagically generated each week)
(so at least any VALID EXISTING official address is accepted)
What I found now (after the migration done by the administrators of the central seat) is that in mailman3:
accept_these_nonmembers is empty
there are 189 entries listed in Non-members (which look those formerly in accept_these_nonmembers)
hold_these_nonmembers was copied identical with the same regexp's as in mailman2
I have inspected a few of them and found that they have Administration options Moderation set to List default
the effect of this apparently was that a post from some.guy@inaf.it (whose subscription address is guy@somewhere.inaf.it) was hold for moderation
is this due to a loop with the regexp in hold_these_nonmembers ?
the default action would be to let it pass
Or should we set Administration options Moderation to Accept immediately ?
if so, how can we do it automatically for all 189 entries ?
or move back the 189 addresses to accept_these_nonmembers ?
or for the future have persons posting with alternate addrresses accepted for the future WITH ONE CLICk ?
AND LAST BUT NOT LEAST would the syntax @list1 in accept_these_nonmembers work in mailman3 ?
already putting it back will save a lot of work.
Many thanks
-- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
"All that is google does not glitter Nor all who use alpine/procmail are lost"