• Andrew C Aitchison via Exim-dev [2025-09-15 20:28]:
On Mon, 15 Sep 2025, Exim Bugzilla via Exim-dev wrote:
https://bugs.exim.org/show_bug.cgi?id=3158
--- Comment #2 from [email protected] ---
(In reply to Jeremy Harris from comment #1)
I assume this is the one you were suggesting "AI" could fix.
Yes and no – same bug I had mentioned on IRC, but I have enough human
intelligence to *never* suggest that so-called "AI" can fix anything.
I was
just experimenting, the day before such experiments helped me
identify one
specific line which caused failed assertion and core dump in mandoc.
So when I
got an Exim core, I was enthusiastic to threw it in as well.
Can you get a debug log, starting at least as far back as the start of
routing?
No, I am afraid I cannot, as those errors are happening very
infrequently.
What is comment, though, is that they occur on cutthrough delivery,
when sending host is *.google.com and destination is also *.google.com.
IIUC it is possible to enable debugging within an ACL.
It sounds as though you have pinned this down well enough
to enable debugging pretty much only on the problem connections.
thank you for this hint! for now I have been running -d -bd, logging
output to a file.
Assuming that you are not Google, I find it odd that you are using
cutthrough on these messages though.
ah, we have been discussing this already:
- https://lists.exim.org/lurker/message/20241216.195455.25c82c6a.en.html
- https://lists.exim.org/lurker/message/20241216.215629.4a2e19cd.en.html
- https://lists.exim.org/lurker/message/20241216.230233.3927272f.en.html
in my case I have 4 users who forward their mail to gmail, and just want
to make sure that G accepts the message before I accept it
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-dev.lists.exim.org/
## unsubscribe (doesn't require an account):
## [email protected]
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://wiki.exim.org/
