You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When doing backups of a master server that are going to be used for slave restore and replication (eg. CONFIG_mysql_dump_master_data=2), and where 1 or more of the dbs on the master server are large, there is the possibility that the binlog file/pos values may be different between the dbs due to --single-transaction happening on a per db basis and time passes between the 1st db being done and the others. Here is an example:
Backups are set to start at 3am
CONFIG_mysql_dump_single_transaction='yes'
CONFIG_mysql_dump_master_data=2
After backup, and when checking data/time of backup files and binlog file/pos:
Aug 21 05:18 db1.sql.gz (large) - mysql-bin.009304/200746006
Aug 21 05:18 db2.sql.gz (small) - mysq-bin.006306/978513887
Aug 21 05:18 db3.sql.gz (small) - mysq-bin.006306/978513887
This makes sense because binlog file/pos is taken at start of backup and time passes between the 1st backup and the other 2. Is there a way to use automysqlbackup for slave restores where the dbs are large and keep same binlog file/pos across all dbs?
The only other option is to do manual backups with 'flush tables with read lock' enabled.
Regards, Robby
The text was updated successfully, but these errors were encountered:
Thank you for your patience. What would you suggest? Saving the binlog position for each database? I don't know if it's a good idea to have a per database dump when setting up a slave server. How would you solve that?
When doing backups of a master server that are going to be used for slave restore and replication (eg. CONFIG_mysql_dump_master_data=2), and where 1 or more of the dbs on the master server are large, there is the possibility that the binlog file/pos values may be different between the dbs due to --single-transaction happening on a per db basis and time passes between the 1st db being done and the others. Here is an example:
After backup, and when checking data/time of backup files and binlog file/pos:
Aug 21 05:18 db1.sql.gz (large) - mysql-bin.009304/200746006
Aug 21 05:18 db2.sql.gz (small) - mysq-bin.006306/978513887
Aug 21 05:18 db3.sql.gz (small) - mysq-bin.006306/978513887
This makes sense because binlog file/pos is taken at start of backup and time passes between the 1st backup and the other 2. Is there a way to use automysqlbackup for slave restores where the dbs are large and keep same binlog file/pos across all dbs?
The only other option is to do manual backups with 'flush tables with read lock' enabled.
Regards, Robby
The text was updated successfully, but these errors were encountered: