On 2/10/25 13:45, Lieuallen, Thomas Otis via Mailman-users wrote:
Mark,
Actually, this is saying cache_life under mailman:
https://docs.mailman3.org/projects/mailman/en/latest/src/mailman/config/docs...
There is a cache_lifetime under ‘dmarc’.
You are correct. I was confused by the dmarc one.
I used your clue about the database to concoct my own solution… 😊. I edited the ‘expires_on’ values in the file_cache table in the database so the files were expired.
It looks like the times in that table are UTC.
I’m going to ratchet down my cache_life so anyone that is playing with templates doesn’t go bonkers (like me) with why their changes aren’t honored. I have a small site, so not much performance impact.
That really shouldn't be necessary, but I think there is an issue in the code at https://gitlab.com/mailman/mailman/-/blob/master/src/mailman/model/template.... which is intended to evict the cache entry for the previous template when a new one is set. If the previous template has a different uri from the one being set, e.g., a mailman:// uri is being replaced by a http(s):// Postorius uri the code will attempt to evict the Postorius uri which is not the cached value.
This should not be an issue when just updating a template already in
Postorius and should not be an issue when setting a Postorius template
to replace one previously the default or in Mailman's var/templates
directory as these normally don't have template table entries, but prior
to Mailman 3.3.6, the mailman import21 command would create template
table entries for imported lists. This was
https://gitlab.com/mailman/mailman/-/issues/988 fixed by
https://gitlab.com/mailman/mailman/-/merge_requests/989
Thank you for your help.
You're welcome. I'm glad you solved it.