I did an upgrade on my mediawiki box the other day to fix some security issues. Well, in the mix, mysql-common was upgraded from mysql-common_5.1.62-1 to mysql-common 5.5.23+dfsg-1. Nothing happened until last night when we took a power hit due to a storm, and when it rebooted, I got
May 25 09:49:06 sisko mysqld[31214]: 120525 9:49:06 [ERROR] /usr/sbin/mysqld: unknown variable 'lc-messages-dir=/usr/share/mysql' May 25 09:49:06 sisko mysqld_safe[31217]: ended May 25 09:49:20 sisko /etc/init.d/mysql[31383]: 0 processes alive and '/usr/bin/mysqladmin --defaults-file=/etc/mysql/debian.cnf ping' resulted in May 25 09:49:20 sisko /etc/init.d/mysql[31383]: /usr/bin/mysqladmin: connect to server at 'localhost' failed May 25 09:49:20 sisko /etc/init.d/mysql[31383]: error: 'Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)' May 25 09:49:20 sisko /etc/init.d/mysql[31383]: Check that mysqld is running and that the socket: '/var/run/mysqld/mysqld.sock' exists! May 25 09:49:20 sisko /etc/init.d/mysql[31383]: May 25 09:57:10 sisko mysqld_safe[10065]: started May 25 09:57:10 sisko mysqld[10068]: 120525 9:57:10 [ERROR] /usr/sbin/mysqld: unknown variable 'lc-messages-dir=/usr/share/mysql' May 25 09:57:10 sisko mysqld_safe[10071]: ended I tried creating mysql.sock, but it just wouldn't start. So I rolled back to 5.1.62, and it (and mediawiki and redmine) came up just fine. Is there some kind of prep work I need to do rather than just springing an unexpected mysql upgrade on a box, or is this bug worthy? I checked for a similar bug on bugs.d.o, but didn't see anything... Thoughts? Thanks, --b -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/cakmzw+asswmhzhmwqjjqroyrvvls8yz_siiarzv+8tbcboh...@mail.gmail.com
