During a recent power hit, our master MySQL database went down. It came back up, but the slaves did not recover the synchronization. One slave is local and also went down in the power hit, the other is remote and didn’t ever go down. Both, however, had the same symptoms – their master log file was that of the master at the point of power failure. After some Googling and finding several potential fixes, I found a new one, and it worked perfectly. On both slaves, I simply hopped on the MySQL CLI and typed:
MYSQL> stop slave; SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1; START SLAVE;
They sync’d right up.