19.1.3.3 GTID Auto-Positioning

GTIDs replace the file-offset pairs previously required to determine points for starting, stopping, or resuming the flow of data between source and replica. When GTIDs are in use, all the information that the replica needs for synchronizing with the source is obtained directly from the replication data stream.

To start a replica using GTID-based replication, you need to enable the SOURCE_AUTO_POSITION option in the CHANGE REPLICATION SOURCE TO statement. The alternative SOURCE_LOG_FILE and SOURCE_LOG_POS options specify the name of the log file and the starting position within the file, but with GTIDs the replica does not need this nonlocal data. For full instructions to configure and start sources and replicas using GTID-based replication, see Section 19.1.3.4, “Setting Up Replication Using GTIDs”.

The SOURCE_AUTO_POSITION option is disabled by default. If multi-source replication is enabled on the replica, you need to set the option for each applicable replication channel. Disabling the SOURCE_AUTO_POSITION option again causes the replica to revert to file-based replication; this means that, when GTID_ONLY=ON, some positions may be marked as invalid, in which case you must also specify both SOURCE_LOG_FILE and SOURCE_LOG_POS when disabling SOURCE_AUTO_POSITION.

When a replica has GTIDs enabled (GTID_MODE=ON, ON_PERMISSIVE, or OFF_PERMISSIVE ) and the SOURCE_AUTO_POSITION option enabled, auto-positioning is activated for connection to the source. The source must have GTID_MODE=ON set in order for the connection to succeed. In the initial handshake, the replica sends a GTID set containing the transactions that it has already received, committed, or both. This GTID set is equal to the union of the set of GTIDs in the gtid_executed system variable (@@GLOBAL.gtid_executed), and the set of GTIDs recorded in the Performance Schema replication_connection_status table as received transactions (the result of the statement SELECT RECEIVED_TRANSACTION_SET FROM PERFORMANCE_SCHEMA.replication_connection_status).

The source responds by sending all transactions recorded in its binary log whose GTID is not included in the GTID set sent by the replica. To do this, the source first identifies the appropriate binary log file to begin working with, by checking the Previous_gtids_log_event in the header of each of its binary log files, starting with the most recent. When the source finds the first Previous_gtids_log_event which contains no transactions that the replica is missing, it begins with that binary log file. This method is efficient and only takes a significant amount of time if the replica is behind the source by a large number of binary log files. The source then reads the transactions in that binary log file and subsequent files up to the current one, sending the transactions with GTIDs that the replica is missing, and skipping the transactions that were in the GTID set sent by the replica. The elapsed time until the replica receives the first missing transaction depends on its offset in the binary log file. This exchange ensures that the source only sends the transactions with a GTID that the replica has not already received or committed. If the replica receives transactions from more than one source, as in the case of a diamond topology, the auto-skip function ensures that the transactions are not applied twice.

If any of the transactions that should be sent by the source have been purged from the source's binary log, or added to the set of GTIDs in the gtid_purged system variable by another method, the source sends the error ER_SOURCE_HAS_PURGED_REQUIRED_GTIDS to the replica, and replication does not start. The GTIDs of the missing purged transactions are identified and listed in the source's error log in the warning message ER_FOUND_MISSING_GTIDS. The replica cannot recover automatically from this error because parts of the transaction history that are needed to catch up with the source have been purged. Attempting to reconnect without the SOURCE_AUTO_POSITION option enabled only results in the loss of the purged transactions on the replica. The correct approach to recover from this situation is for the replica to replicate the missing transactions listed in the ER_FOUND_MISSING_GTIDS message from another source, or for the replica to be replaced by a new replica created from a more recent backup. Consider revising the binary log expiration period (binlog_expire_logs_seconds) on the source to ensure that the situation does not occur again.

If during the exchange of transactions it is found that the replica has received or committed transactions with the source's UUID in the GTID, but the source itself does not have a record of them, the source sends the error ER_REPLICA_HAS_MORE_GTIDS_THAN_SOURCE to the replica and replication does not start. This situation can occur if a source that does not have sync_binlog=1 set experiences a power failure or operating system crash, and loses committed transactions that have not yet been synchronized to the binary log file, but have been received by the replica. The source and replica can diverge if any clients commit transactions on the source after it is restarted, which can lead to the situation where the source and replica are using the same GTID for different transactions. The correct approach to recover from this situation is to check manually whether the source and replica have diverged. If the same GTID is now in use for different transactions, you either need to perform manual conflict resolution for individual transactions as required, or remove either the source or the replica from the replication topology. If the issue is only missing transactions on the source, you can make the source into a replica instead, allow it to catch up with the other servers in the replication topology, and then make it a source again if needed.

For a multi-source replica in a diamond topology (where the replica replicates from two or more sources, which in turn replicate from a common source), when GTID-based replication is in use, ensure that any replication filters or other channel configuration are identical on all channels on the multi-source replica. With GTID-based replication, filters are applied only to the transaction data, and GTIDs are not filtered out. This happens so that a replica’s GTID set stays consistent with the source’s, meaning GTID auto-positioning can be used without re-acquiring filtered out transactions each time. In the case where the downstream replica is multi-source and receives the same transaction from multiple sources in a diamond topology, the downstream replica now has multiple versions of the transaction, and the result depends on which channel applies the transaction first. The second channel to attempt it skips the transaction using GTID auto-skip, because the transaction’s GTID was added to the gtid_executed set by the first channel. With identical filtering on the channels, there is no problem because all versions of the transaction contain the same data, so the results are the same. However, with different filtering on the channels, the database can become inconsistent and replication can hang.