Description of problem:

I just found that /etc/systemd/system.conf was having this:
-------- 8< ---------------- 8< ---------------- 8< ---------------- 8< --------
DefaultLimitCORE=0:infinity
-------- 8< ---------------- 8< ---------------- 8< ---------------- 8< --------

Thanks to this "feature", we don't get any coredump when a service fails, which is not that optimal for our support work.
Please drop this patch, it's a regression compared to RHEL8.

Version-Release number of selected component (if applicable):

systemd-252

How reproducible:

Always

Steps to Reproduce:
1. Kill sshd with SEGV signal

1. kill -11 $(cat /run/sshd.pid)

Actual results:

No coredump created:
-------- 8< ---------------- 8< ---------------- 8< ---------------- 8< --------

1. journalctl -u systemd-coredump@0-2859-0.service
   Jun 07 12:00:41 vm-uefi9 systemd[1]: Started Process Core Dump (PID 2859/UID 0).
   Jun 07 12:00:41 vm-uefi9 systemd-coredump[2860]: Resource limits disable core dumping for process 782 (sshd).
   Jun 07 12:00:41 vm-uefi9 systemd-coredump[2860]: [🡕] Process 782 (sshd) of user 0 dumped core.
   Jun 07 12:00:41 vm-uefi9 systemd[1]: systemd-coredump@0-2859-0.service: Deactivated successfully.

1. coredumpctl
   TIME PID UID GID SIG COREFILE EXE SIZE
   Wed 2023-06-07 12:00:41 CEST 782 0 0 SIGSEGV none /usr/sbin/sshd -
   • • • • • • • 8< ---------------- 8< ---------------- 8< ---------------- 8< --------

Expected results:

Coredump being created

Additional info:

RHEL8 behaviour is correct:
-------- 8< ---------------- 8< ---------------- 8< ---------------- 8< --------

1. systemctl show | grep CORE
   DefaultLimitCORE=infinity
   DefaultLimitCORESoft=infinity
   • • • • • • • 8< ---------------- 8< ---------------- 8< ---------------- 8< --------

Being able to analyze core dumps is critical to RHEL Support.