replication-manager version 1.1 is released in a stable version. This release contains long-awaited changes such as support for MySQL 5.7, Binlog Flashback, multi-cluster mode and various stability fixes.
The following features have been added to the Replication-Manager 1.1 Stable Release:
replication-manager is now able to manage multiple database clusters. No change is needed except declaring multiple sections in the configuration file such as
Declared clusters in the configuration file will be monitored by default, but you can also specify them optionally with a flag:
Similar to the multi cluster feature,
replication-manager is now able to handle multiple proxies (MaxScale, ProxySQL, HAProxy, ShardProxy...)
Rejoin of failed nodes has been refactored and widely improved. It is capable to use binary log backups, Flashback (with the supplied mysqlbinlog binary) or mysqldump to perform lossless node rejoin.
replication-manager can now be used in active/standby mode for HA deployments. This requires the installation of a small daemon otherwise known as
arbitrator that uses almost no resources save for a TCP port and SQLite backend to save states.
replication-manager now embeds a lightweight graphite server that can be turned on at runtime to perform metric collection and graph display.
False positive detection has been improved with the support of replication heartbeat, external http calls and maxscale monitor.
Multi-tier topologies (for example Master->Intermediary Master->Slave) are now supported and correctly detected.
Grab a release on GitHub
Replication Manager for MySQL and MariaDB is an open-source platform for handling MySQL, MariaDB and Percona Server replication topologies. It is developed by experts in consulting and remote database services.
Replication Manager for MySQL and MariaDB is commercially supported by Signal 18. Contact us for any support requests.
If you would like to report a bug, ask for a feature or have questions, open an issue on Github