Similar problem here. My apache2 server goes into a reload loop every
time /etc/logrotate.d/apache2 is called. That is, daily a few minutes
after midnight. The loop is triggered by the reload statement in the
logrotate script.
As a band-aid I just replaced "reload" by "restart" in the logrotate script.
In my case kern.log attributes the segfault to libglib-2.0:
# tail -n 1 /var/log/kern.log
Jul 24 02:22:39 bibo kernel: [175623.347801] /usr/sbin/apach[8662]: segfault at 7fb516473660 ip 00007fb516473660 sp 00007ffe8b10e508 error 14 in libglib-2.0.so.0.5600.1[7fb516dac000+113000]
If I apply the /proc/$pid/maps|sort search with ldd I get:
# pid=770; for i in $(awk '{ print $6 }' < /proc/$pid/maps|sort -u|grep /) ; do ldd -d $i|grep libglib && echo $i ; done
ldd: /dev/zero: not regular file
ldd: /SYSV6405bf18: No such file or directory
libglib-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0 (0x00007fcfb3308000)
/usr/lib/php/20170718/imagick.so
libglib-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0 (0x00007f0a395be000)
/usr/lib/x86_64-linux-gnu/liblqr-1.so.0.3.2
libglib-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0 (0x00007f2b15ea5000)
/usr/lib/x86_64-linux-gnu/libMagickCore-6.Q16.so.5.0.0
libglib-2.0.so.0 => /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0 (0x00007fa7bdb1c000)
/usr/lib/x86_64-linux-gnu/libMagickWand-6.Q16.so.5.0.0
---<)kaimartin(>---
--
Kai-Martin Knaak
Email: kmk@familieknaak.de
Öffentlicher PGP-Schlüssel:
https://keyserver.ubuntu.com/pks/lookup?op=index&search=0x7B0F9882
Attachment:
pgpYDg2ETpN5E.pgp
Description: OpenPGP digital signature