On Fri, 21 Feb 2020 at 01:16, Tom Horsley <horsley1953@gmail.com> wrote:
On Fri, 21 Feb 2020 00:55:00 +0000
Anthony F McInerney wrote:

> I guess at this point
> uname -a
> systemctl --version
> cat /proc/cmdline
> Would be interesting. (Apologies if you posted this or attached it to the
> bug report - i don't appear to see anything for these)

Not sure why it would be interesting, but it is easy to get:

[root@zooty ~]# uname -a
Linux zooty.my.lan 5.4.19-200.fc31.x86_64 #1 SMP Wed Feb 12 15:21:24 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux
[root@zooty ~]# systemctl --version
systemd 243 (v243.7-1.fc31)
+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD +IDN2 -IDN +PCRE2 default-hierarchy=unified
[root@zooty ~]# cat /proc/cmdline
BOOT_IMAGE=(hd0,msdos5)/boot/vmlinuz-5.4.19-200.fc31.x86_64 root=UUID=47a0b43c-2a9d-4929-81e5-860ece806911 ro selinux=0 audit=0 net.ifnames=0 biosdevname=0 rd.driver.blacklist=nouveau modprobe.blacklist=nouveau nvidia-drm.modeset=1

What would really be interesting is a way to get a backtrace
from systemd when it is waiting for the service to see what
in the sam hill made it decide to wait :-).

On the bug report, you ran dhcp -f (without using systemd?),same issue, SIGINT being ignored, and again with httpd (with or without systemd i guess it doesn't matter).

So it seems, that SIGINT is also being ignored when systemd sends it to the process? (And then you ran into the infamous 1 min 30 sec+ timer, since the process won't exit gracefully?).

So rather,  you have some odd, kernel issue maybe?

Lotta questions marks there, as it seems you still think the issue is sytemd, and i'm wondering if i misread something...