2012/1/24 Suporte Avanutri <suporte2@stripped>:
> I've tried this before, but the server stills going down. The first error is always
> this:
>
> Couldn't execute 'SELECT /*!40001 SQL_NO_CACHE */* FROM 'usuario': Lost connection to
> MySQL server during query (2013)
What's likely happening here is that the access to the table is
causing a fatal error inside your mysql daemon, it dies and restarts,
resulting in the log messages you see about mysql not being shut down
properly and doing a crash recovery. The followup suggestion was for
you to put "innodb_force_recovery=1" in your my.cnf and restart it,
but that didn't work for you. The value that you pass to
innodb_force_recovery controls how much of the innodb startup / crash
recovery process to use. The gory details are at
http://dev.mysql.com/doc/refman/5.0/en/forcing-innodb-recovery.html .
Valid values are between 1 and 6.
For example, I shut down my mysql daemon, I make a copy of my
corrupted database at /var/lib/mysql/ into /var/lib/mysql_tmp/, and
then manually start the mysql daemon in the foreground with this:
su - mysql -c '/usr/libexec/mysqld --innodb_force_recovery=6 --datadir
/var/lib/mysql_tmp/'
It simply sounds like you need to find the appropriate value to use to
make your data accessible.
Regards... Todd
--
SOPA: Any attempt to [use legal means to] reverse technological
advances is doomed. --Leo Leporte
