September 1, 2014

How to create/restore a slave using GTID replication in MySQL 5.6

MySQL 5.6 is GA! Now we have new things to play with and in my personal opinion the most interesting one is the new Global Transaction ID (GTID) support in replication. This post is not an explanation of what is GTID and how it works internally because there are many documents about that:

http://dev.mysql.com/doc/refman/5.6/en/replication-gtids-concepts.html

One thing that worths to mention is that if you want GTID support log_slave_updates will need to be enabled in slave server and the performance impact should be taken in account.

Anyway, this post tends to be more practical, we will see how to create/restore new slaves from a master using GTID.

How to set up a new slave

The first thing that we need to know is that now Binary Logs and Position are not needed anymore with GTID enabled. Instead we need to know in which GTID is the master and set it on the slave. MySQL keeps two global variables with GTID numbers on it:

gtid_executed: it contains a representation of the set of all transaction logged in the binary log
gtid_purged: it contains a representation of the set of all transactions deleted from the binary log

So now, the process is the following:

    • take a backup from the master and store the value of gtid_executed
    • restore the backup on the slave and set gtid_purged with the value of gtid_executed from the master

    The new mysqldump can do those tasks for us. Let’s see an example of how to take a backup from the master and restore it on the slave to set up a new replication server.

    Now we take a backup with mysqldump from the master:

    It will contain the following line:

    Therefore during the dump recover process on the slave it will set GTID_PURGED to the GTID_EXECUTED value from the master.

    So now, we just need to recover the dump and start the replication:

    The last step is to configure the slave using the auto-configuration method of GTID:

    How to restore a slave in a bad and fast way

    Let’s imagine that our slave has been down for several days and the binary logs from the master have been purged. This is the error we are going to get:

    So, let’s try to solve it. First we have the bad and fast way, that is, point to another GTID that the master has in the binary logs. First, we get the GTID_EXECUTED from the master:

    And we set it on the slave:

    Error! Remember, we get the GTID_EXECUTED from the master and set is as GTID_PURGED on the slave.

    Error again, GTID_EXECUTED should be empty before changing GTID_PURGED manually but we can’t change it with SET because is a read only variable. The only way to change it is with reset master (yes, on a slave server):

    Now, if you don’t get any error like primary/unique key duplication then you can run the pt-table-checksum and pt-table-sync.

    How to restore a slave in a good and slow way

    The good way is mysqldump again. We take a dump from the master like we saw before and try to restore it on the slave:

    Wop! It is important to mention that these kind of error messages can dissapear on the shell buffer because the restore of the dump will continue. Be cautious.

    Same problem again so same solution too:

    Conclusion

    With the new GTID we need to change our minds. Now binary log and position is not something we need to take in account, gtid_executed and gtid_purged are our new friends. Newer versions of Xtrabackup have full support of GTID. You can check the following blog post:

    http://www.percona.com/blog/2013/05/09/how-to-create-a-new-or-repair-a-broken-gtid-based-slave-with-percona-xtrabackup/

  • About Miguel Angel Nieto

    Miguel joined Percona in October 2011. He has worked as a System Administrator for a Free Software consultant and in the supporting area of the biggest hosting company in Spain. His current focus is improving MySQL and helping the community of Free Software to grow. Miguel's roles inside Percona are Senior Support Engineer and Manager of EMEA Support Team.

    Comments

    1. Hi,
      I thought the whole idea was for slaves *not* needing to know the GTID in advance, but rather negotiate it with the server.
      For example, I set a replication setup via mysqlsandbox, had some replicating statements going, switched to gtid, had some further statements going around; stopped slave, RESET SLAVE, reconnected slave via CHANGE MASTER TO, but only with credentials (host, port, user, password), followed by START SLAVE. So it found out the gtid by itself and resumed from the right place.
      So, again, when is a need to pass around the GTID_PURGED from master to slave?

    2. Fernando says:

      As I see, to do replicas, percona xtrabackup still be better than this when we’re talking about large databases.

    3. Shlomi, yes, if everything goes as expected then you don’t need to play with GTID_PURGED or GTID_EXECUTED. That’s the reason I forced a problem in my blog post:

      “Last_IO_Error: Got fatal error 1236 from master when reading data from binary log: ‘The slave is connecting using CHANGE MASTER TO MASTER_AUTO_POSITION = 1, but the master has purged binary logs containing GTIDs that the slave requires.’”

      That can happen often if you purge binary logs from the master and then some manual process and dealing with GTID_PURGED is needed. Usually tools like mysqldump do this for you, but as you can see in the last example, it can also fail.

    4. marc castrovinci says:

      “One thing that worths to mention is that if you want GTID support log_slave_updates will need to be enabled in slave server and the performance impact should be taken in account.”

      Is the performance impact any different then using log_slave_updates as it currently stands in 5.5?

    5. Marc, no, the same impact. The only difference is that with GTID log_slave_updates it’s mandatory.

    6. marc castrovinci says:

      Also, any idea how this process would work using xtrabackup? I’m assuming there would be an option to dump the GTID_PURGED or GTID_EXECUTED into a file?

    7. marc, that’s the idea. Create a file in the backup with the content of GTID_EXECUTED. But it is still something we have to decide and test.

    8. Is there any way to just reset the replication for 1 specific table that has LOTS of skipped/bad transactions?

    9. Lalit says:

      mysql 5.6 GTID replication error

      I recently upgraded from mysql 5.5 to 5.6.15 and enabled GTID based replication, now problem is for some reasons we restarted our slave and after that replication break.

      I Checked SHOW SLAVE STATUS error is “‘Got fatal error 1236 from master when reading data from binary log: ”The slave is connecting using CHANGE MASTER TO MASTER_AUTO_POSITION = 1, but the master has purged binary logs containing GTIDs that the slave requires.”

      i am suspecting it happens because of master old binary log (7 days before) deleted form master as we are using expire_logs_days=7 and slave still looking for this old binary log for some GTID reference to replicate with master.
      I tried following link solution ,
      http://www.mysqlperformanceblog.com/…-in-mysql-5-6/
      but it still not working.
      Note :using mysql 5.6.15

      Now only option for me to re-create slave by taking mysqldump from master.

      Is there any other option if we can start replication without recreating new slave.

      Thanks in Advance

    10. shahid says:

      Hi ,
      I m getting one error and unable to trace it.
      In my master server i have disabled master-verify-checksum but still show binlog event not ignoring it..
      mysql> show global variables like ‘%check%’;
      +—————————+——–+
      | Variable_name | Value |
      +—————————+——–+
      | binlog_checksum | NONE |
      | foreign_key_checks | ON |
      | innodb_checksum_algorithm | innodb |
      | innodb_checksums | ON |
      | innodb_file_format_check | ON |
      | master_verify_checksum | OFF |
      | slave_checkpoint_group | 512 |
      | slave_checkpoint_period | 300 |
      | slave_sql_verify_checksum | OFF |
      | unique_checks | ON |
      mysql> show binlog events in ‘mysql225-bin.000002′;
      ERROR 1220 (HY000): Error when executing command SHOW BINLOG EVENTS: Wrong offset or I/O error

    Speak Your Mind

    *