Dear Maintainer,
I keep finding many messages like this one in paniclog:

2024-01-11 09:55:20 1rNqqH-008cnX-0s SIGSEGV (fault address: 0x58)
2024-01-11 09:55:20 1rNqqH-008cnX-0s SEGV_MAPERR
2024-01-11 09:55:20 1rNqqH-008cnX-0s SIGSEGV (null pointer indirection)
2024-01-11 09:55:20 1rNqqH-008cnX-0s SIGSEGV (2055771 delivering 1rNqqH-08cnX-0s: just tried mx01.mail.icloud.com [17.57.152.5] for username@me.com: result OK
)
2024-01-11 09:55:20 1rNqqH-008cnX-0s Delivery status for username@me.com: got 0 of 7 bytes (pipeheader) from transport process 2055771 for transport smtp


This happens when I receive a message that should be stored locally in Maildir
format and also forwarded to a remote smtp that deny the delivery. In the
following log you see Apple remote smtp server denying the delivery because
of a Spamhaus list.

root@mail:~# fgrep 1rNqqH-008cnX-0s /var/log/exim4/mainlog
2024-01-11 09:55:14 1rNqqH-008cnX-0s <= 10694-91-78766-814-myusername=mydomain.com@mail.angelomathews.life H=lodha.lunchissez.com (yellow.angelomathews.life) [157.52.182.152] P=esmtp K S=6890 id=x6ofxefs3bzzjj53-vn0k0sre12sqfx87-133ae@angelomathews.life
2024-01-11 09:55:14 1rNqqH-008cnX-0s => myusername <myusername@mydomain.com> R=real_local_from_anywhere T=maildir_home
2024-01-11 09:55:17 1rNqqH-008cnX-0s H=mx01.mail.icloud.com [17.56.9.29]: SMTP error from remote mail server after RCPT TO:<myusername@me.com>: 550 5.7.1 Your email was rejected due to having a domain present in the Spamhaus DBL -- see https://www.spamhaus.org/dbl/
2024-01-11 09:55:20 1rNqqH-008cnX-0s SIGSEGV (fault address: 0x58)
2024-01-11 09:55:20 1rNqqH-008cnX-0s SEGV_MAPERR
2024-01-11 09:55:20 1rNqqH-008cnX-0s SIGSEGV (null pointer indirection)
2024-01-11 09:55:20 1rNqqH-008cnX-0s SIGSEGV (2055771 delivering 1rNqqH-008cnX-0s: just tried mx01.mail.icloud.com [17.57.152.5] for myusername@me.com: result OK
2024-01-11 09:55:20 1rNqqH-008cnX-0s Delivery status for myusername@me.com: got 0 of 7 bytes (pipeheader) from transport process 2055771 for transport smtp
2024-01-11 09:55:20 1rNqqH-008cnX-0s == myusername@me.com <myusername@mydomain.com> R=dnslookup T=remote_smtp defer (-1): smtp transport process returned non-zero status 0x000b: terminated by signal 11
2024-01-11 09:55:20 1rNqqO-008cng-0b <= <> R=1rNqqH-008cnX-0s U=Debian-exim P=local S=810
2024-01-11 09:55:20 1rNqqH-008cnX-0s Frozen
2024-01-11 09:56:01 1rNqqH-008cnX-0s removed by root
2024-01-11 09:56:01 1rNqqH-008cnX-0s Completed

Thank you,
Giuseppe

[...]

Is this a regression in 4.96-15+deb12u4? i.e. can you still reprroduce
the issue after downgrading exim4-daemon-heavy to 4.96-15+deb12u3 [1]
(and re-reproduce after upgrading again)?

cu Andreas

[1] http://ftp.debian.org/debian/pool/main/e/exim4/exim4-daemon-heavy_4.96-15+deb12u3_amd64.deb

Hello Andreas,

Il giorno ven, 12/01/2024 alle 18.15 +0100, Andreas Metzler ha scritto:
[...]

I am going to downgrade the package and let it run for a week. I'll report
here.

Thank you,
Giuseppe

This does not seem to be a regression; I tried as far back as deb12u2 and
the same issue could be reproduced.

Best,
OdyX

Control: found 1060457 4.96-15+deb12u2
[...]

Thank you.

Would you be able to check with 4.97, too? I intend to provide backports
once testing has stopped being stuck due to t64.

cu Andreas

No; it's a production system, and downgrading along the past stable
releases was OK'ish; upgrading to unstable's, less so.

Just to report a successful upgrade:

I had exactly the same issue on one of my mail servers, with exim
versions up to 4.96-15+deb12u7. It was even triggered by the the same recipient
domains (icloud.com and me.com).

Upgrading to 4.98.1-2~bpo12+1 from bookworm-backports solved the issue.
(Now I get a proper error message, "SMTP error from remote mail server
after RCPT TO", because it looks like my IP is on some blocklist.)

In case it's relevant: This was on a rather old (and often upgraded)
system that is still using the i386 architecture.

Jan