Am 23.12.24 um 11:17 schrieb Jeremy Harris via Exim-users:
I'd be tending to blame systemd, also. What has it started doing differently?
Debian 11 worked fine without any problems. The problems started with Debian 12 and the backported version. It is obvious that the systemd files changed and now the old init script is not used anymore. It "feels" like a timeout and systemd finally killing the daemon. But I have no facts to support this at the moment.
Can it be reverted?
This is the next step on the agenda. Going back to Debian 12 Exim 4.96 would mean to reintroduce the Google <-> ARC errors I encountered but having a 100% working mail delivery system is more important.
With regard to the long restart time: if you just "kill -HUP" the PID of the exim daemon, watching the log, how long is it before the new deamon is reported as running?
SIGHUP works immediately.
Does the systemd config support a "reload" operation? How long does that take?
Same as the SIGHUP.
Presumably a "restart" is doing a "stop" first, so it's the time there that matters. What does "exiwhat" report during that time period?
It took two minutes to restart and exiwhat only showed this in the meantime (I repeated the exiwhat several times): $: exiwhat 2700875 handling TLS incoming connection from xmbghk7.mail.qq.com [43.163.128.54]
smime.p7s
Description: Kryptografische S/MIME-Signatur
-- ## subscription configuration (requires account): ## https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exim.org/ ## unsubscribe (doesn't require an account): ## exim-users-unsubscr...@lists.exim.org ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://wiki.exim.org/
