So I upgraded one of my MTA's this morning to 6.0.9, and for some reason the upgrade was getting stuck on Starting Mysql. Everything I've read points to a corrupt database.
Everything I found points to the Mysql Crash Recovery
When I get to this step:
mysql --batch --skip-column-names -e "show databases" | grep -e mbox -e zimbra > /tmp/mysql.db.list
I get the error below:
[ERROR] Fatal error: Can't open and lock privilege tables: Table 'mysql.host' doesn't exist
I've got a call into support with a server down, but just wanted to reachout to the forums to see if anyone can help as well...
If your MySQL Server is not starting you may try to repair the table by
This can fix almost anything on an MyISAM table.
/opt/zimbra/mysql/bin/myisamchk -r /opt/zimbra/db/data/mysql/*.MYI
So it turned out that I accidently rm -rf'd the SQL logs...it's was all bad. Even though I ctrl+c'd very quickly when I saw what I had typed, I had to work with support on rebuilding the mysql tables.
how it was done? zmsetup.pl and choose to only install mysql?
... that will be not necessary in most cases. If you only deleted the InnoDB transaction logs
Originally Posted by malayo
than stopping Zimbra via "zmcontrol stop" is. O.K. As second step only start and stop the Database (as the zimbra user)
if nothing else is corrupted, restarting the MySQL Server will recreate the InnoDB transaction logs. If this doesn't lead to success, there's something else.
mysql> show innidb status\G
thanks, apparently my problem of mysql.server and mailboxd would not stated as running was causing by not using right mysql password after copying localconfig.xml from old server for migration purposes (centos 5 to centos 6 because zcs 8 only runs on centos 6)