Apache dies every night since upgrading to Debian 8

[H9k]

Since upgrading Plesk 12.5.30 from Debian 7 to Debian 8, I have a problem with Apache 2.4.
Every night, during some daily cron I guess, it dies:

[Mon Oct 12 02:11:59.093175 2015] [mpm_worker:notice] [pid 5559:tid 3074590528] AH00297: SIGUSR1 received. Doing graceful restart
[Mon Oct 12 02:12:01.199716 2015] [core:notice] [pid 5559] AH00060: seg fault or similar nasty error detected in the parent process

If I reload apache manually, it works fine. My guess is that something is going wrong with log rotation, but I cannot see what could be the problem.
Has anyone experienced such an issue?

 

[mizar]

Hello,

I remember same problems with server graceful reload in short time, i.e */etc/init.d/apache2 reload && /etc/init.d/apache2 reload && /etc/init.d/apache2 reload* leads to segmentation fault.
You can try to replace reload to restart in /etc/logrotate.d/apache2 to make sure that this problem caused by logrotation

 

[H9k]

Yeah I did that and it at least is a workaround, apache does not crash anymore on logrotate.
I have a feeling that Apache 2.4 is less reliable than 2.2, maybe that has something to do with the fact that it runs in a OpenVZ container which still has the 2.6 kernel. I will upgrade to kernel 4.2 and LXC container sometime over the next weeks/months, until then I hope I won't get any other major problems...