Am 13.11.24 um 12:13 schrieb Jeremy Harris via Exim-users:
> A quick look though git commit logs doesn't throw up anything for ARC for that period.
> We'd probably have to bisect, if you care enough, to locate the exact change.
I'm fine with using 4.98 from the Debian Backports. Locating such a change is in my experience
almost always a PITA and not worth your time.
> Has the 5% fail rate from other senders also cleared up?
No, but right now my best guess would be that those are "legitimate" fails. I've only
implemented this on a staging server which right now handles mailing list traffic,
and I would assume that those mails are somehow mistreated on the way:
arc_state = fail , arc_state_reason = AMS body hash miscompare , arc_domains = microsoft.com , arc_oldest_pass = 0
arc_state = fail , arc_state_reason = AMS body hash miscompare , arc_domains = microsoft.com , arc_oldest_pass = 0
arc_state = fail , arc_state_reason = AMS body hash miscompare , arc_domains = microsoft.com , arc_oldest_pass = 0
I'll be implementing ARC verifying and signing next week on our production systems and
would then check if it's still a problem. But even then a failing verifier wouldn't be that
bad as the "real" test for incoming mails is done in rspamd and I'm only verifying it for
forwarded mails and our ARC signing later on in the chain.
--
## subscription configuration (requires account):
##
https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exim.org/
## unsubscribe (doesn't require an account):
## exim-users-unsubscribe@???
## Exim details at
http://www.exim.org/
## Please use the Wiki with this list -
http://wiki.exim.org/
