OpenVPN process crashes and never recovers, can't be "kill -9"-ed

[mopritz]

Dear Community,

I've encountered a really weird issue that I'm hoping someone here can help me with. On-premises of a customer I have a SG-1100 (2.4.5-RELEASE-p1). Now over the weekend I've noticed that a previously working OpenVPN P2P instance has just ... stopped working. The "status" page looks like this:

And after trying to stop and start the service, it looks like this:

The accompanying logs from system.log look like this when trying to restart it:

May 17 11:31:44 PFSENSE-01 php-fpm[14654]: /rc.start_packages: Restarting/Starting all packages.
May 17 11:32:12 PFSENSE-01 php-fpm[14287]: OpenVPN terminate old pid: 81405
May 17 11:32:15 PFSENSE-01 php-fpm[14287]: /status_services.php: OpenVPN ID server4 PID 81405 still running, killing.
May 17 11:32:16 PFSENSE-01 php-fpm[14287]: /status_services.php: OpenVPN ID server4 PID 81405 still running, killing.
May 17 11:32:17 PFSENSE-01 php-fpm[14287]: /status_services.php: OpenVPN ID server4 PID 81405 still running, killing.
May 17 11:32:17 PFSENSE-01 php-fpm[14287]: /status_services.php: OpenVPN ID server4 PID 81405 still running, killing.
May 17 11:32:18 PFSENSE-01 php-fpm[14287]: /status_services.php: OpenVPN ID server4 PID 81405 still running, killing.
May 17 11:32:19 PFSENSE-01 php-fpm[14287]: /status_services.php: OpenVPN ID server4 PID 81405 still running, killing.

And this goes on forever or until I restart PHP-FPM via SSH.

The process with that PID is indeed still in the process list, but it can't be killed via any method I've tried so far. pkill, kill, kill -9; neither of them do anything (both as admin user and as root). The service keeps running:

root    81405   0.0  0.1   9756  1360  -  Ds   11:31      0:00.01 /usr/local/sbin/openvpn --config /var/etc/openvpn/server4.conf

Any idea why this could be, or what I can do to fix this?

Best Regards,
Mo