[debian-mysql] Bug#830976: logrotate failures if mysql server is not running

[Otto Kekäläinen]

Hello!

2016-12-14 15:42 GMT+02:00 Meik Hellmund <Meik.Hellmund at math.uni-leipzig.de>:
> Followup-For: Bug #830976
>
>
> I got  errors from cron:
>  /etc/cron.daily/logrotate:
>  logrotate_script: 3: [: /var/run/mysqld/mysqld.pid: unexpected operator
>
> The reason I found: The postrotate script of /etc/logrotate.d/mysql-server
> contains
>   my_print_defaults --mysqld | grep -oP "pid-file=\K[^$]+"
> and this generates on my machine two identical lines of output:
> /var/run/mysqld/mysqld.pid
> /var/run/mysqld/mysqld.pid
>
> So I added  " | uniq"  after the grep. Perhaps " | head -n1"
> would be even better.


I seems this was introduced in commit
https://anonscm.debian.org/cgit/pkg-mysql/mariadb-10.0.git/commit/debian/mariadb-server-10.0.mysql-server.logrotate?id=2948deb77b61f805e41b7032c71226a69d29a317
when importing upstream 10.0.25.

The original upstream commit that changed this was
https://github.com/MariaDB/server/commit/8942824a5304e23f765b88d07498786d80092843

I suggest you work directly with upstream to get this syntax fine
tuned and fixed globally.