MySQL failed after update (socket error)

[Holbrookau]

FreeBSD 5.4 dedicated server. Was running cPanel/WHM 10, Apache 1.2, PHP 4.(?), MySQL 4.0.
A couple of days ago I went to install curl, however this required a later verion of PHP. No problem I thought, except that MySQL also needed to be updated before hand.
So I ran Tweak Settings, selected MySQL 4.1, then ran Update Server Software - I didn't realise at the time but MySQL apparently failed at that point and hasn't worked since.
Oblivious to the fact MySQL was dead I updated PHP to v5 and added the curl module. After checking my server emails I realised MySQL was down..
I have spent the last couple of days upgrading, downgrading and sidewaysgrading, all to no avail. I have read and appled most every fix I could find here on the issue but MySQL will not come back - I've managed to kill off Apache a couple of times (but is working now), although SpamAssasion has gone AWOL since my last attempt.
I did find in the logs that pkgdb was not working correctly at the time of the first upgrade - may have something to do with it?
Currenty the server is running cPanel/WHM 11, Apache 1.3.39, PHP 4.4.7 and MySQL 4.0.

Any help to get this fixed will be much appreciated (suppossed to going on holidays in 2 days...).

Update Server Software returns the following:
/usr/local/bin/mysqladmin: connect to server at 'localhost' failed
error: 'Can't connect to local MySQL server through socket '/tmp/mysql.sock' (2)'
Check that mysqld is running and that the socket: '/tmp/mysql.sock' exists!
Unable to reset root MySQL password. at /scripts/securemysql line 154.

Restart SQL Server in WHM shows tihs:
mysql status
/usr/local/bin/mysqladmin: connect to server at 'localhost' failed error: 'Can't connect to local MySQL server through socket '/tmp/mysql.sock' (2)' Check that mysqld is running and that the socket: '/tmp/mysql.sock' exists! mysql has failed, please contact the sysadmin (result was "Warning, no valid mysql.sock file found.mysql has failed").

The server MySQL error log reads:(This is from about the time of the first upgrade to the end of file)
070925 10:08:13 /usr/local/libexec/mysqld: Normal shutdown

070925 10:08:18 InnoDB: Starting shutdown...
070925 10:08:27 InnoDB: Shutdown completed
070925 10:08:27 /usr/local/libexec/mysqld: Shutdown Complete

070925 10:08:27 mysqld ended

070925 10:10:18 mysqld started
070925 10:10:21 InnoDB: Started
mysqld got signal 11;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
We will try our best to scrape up some info that will hopefully help diagnose
the problem, but since we have already crashed, something is definitely wrong
and this may fail.

key_buffer_size=8388600
read_buffer_size=131072
max_used_connections=0
max_connections=500
threads_connected=0
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_connections = 1096188 K
bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

070925 10:10:21 mysqld restarted
070925 10:10:21 InnoDB: Started
/usr/local/libexec/mysqld: ready for connections.
Version: '4.0.24' socket: '/tmp/mysql.sock' port: 3306 FreeBSD port: mysql-server-4.0.24
070925 10:12:52 mysqld restarted
070925 10:12:52 /usr/local/libexec/mysqld: unknown variable 'old-passwords=1'

070925 10:12:52 mysqld ended

070925 10:12:52 mysqld started
070925 10:12:52 /usr/local/libexec/mysqld: unknown variable 'old-passwords=1'

070925 10:12:52 mysqld ended

[serversphere]

Try adding this to /etc/my.cnf:

skip-innodb

You can put it in the [mysqld] section. Also you might try lowering the change key_buffer_size or the max_connections, then restart mysql. See how that works for you.

[Holbrookau]

Thanks for the reply, however.. Altering my.cnf makes no differance, I'd assume that MySQL doesn't even get far enough for the file to be read before it dies - there are no entries in the err file after the first update failed.
I'm away for a week or so as from now so the few casual users I have will have to go without forums until I get back and sort this issue..
Thanks anyway.

[cpanelkenneth]

The pkgdb error is quite likely the culprit. Please open a support ticket at http://tickets.cpanel.net/submit/

If MySQL was installed from ports, via SSH/CLI goto the mysql40-server (I think that is the name) and do a make reinstall

If you are wanting PHP 5, you should skip MySQL 4.1 and go directly MySQL 5.


If your FreeBSD install is up-to-date, you might be able to use binary packages, which would speed up the process some (no need to compile MySQL).

The above are just generalized tips that may or may not work. Opening a support ticket would enable us to get your server in tip-top shape quickly.

[Holbrookau]

Thanks for the reply Kenneth, sorry for the delay but I've just returned from holidays..
I believe I fixed the pkg issue not long after the problems started, however it makes no differance to further attemps at fixing things.
I would be most appreciative if you can look at this - ticket lodged.