replication-manager

In case of MDEV-15174 MariaDB can crash on SQL show plugins use monitoring-plugins = false to disable show plugins

-Orchestrator onpremise fixing multiple ssh script executions

Onpremise ssh script add a custom TLS private key via parameter onpremise-ssh-private-key

• Multi source child cluster using replication filters can break (part 1: fixing switchover via flush binary log after freeze master)
• On demand SSH script better discovering of ssh key as systemd does not expose $HOME

Fixing MariaDB 10.6 slow log capture

• Cluster server state can freeze on suspect
• Tracking unprovision resquest via cookie

• Fixing freeze caused by regression after reload config patch
• Fixing rolling restart
• Fixing spider compliance file

• Fixing state cookies failed to get removed
• Tracking no backups and auto reseed when backup executed
• Split configuration into his own package to enable external tool usage
• Fixing flashback
• Fixing reload config will transite master to read only and let proxysql without master
• Taking into account tag readonmaster for proxysql for initializing server code

• Fixing Flashback failed as always looking for all slave status and wrongly looking at the new master id
• Fixing possible minor crashes

Fixing some flooding warnings on leader database binary log rotation

• Rig the election for switchover in case of multiple prefered master
• Fixing switchover with multiple prefered master #345

• Fixing crash in adding ne cluster from service plan
• Preserving state for server CheckPrivileges when done every minute

Official release of Replication Manager 2.1.1

• Fix reverse DNS issues
• Fix removed slave causing error 1045
• Force slave_pos when MariaDB GTID Switch
• Fix crash in haproxy stats
• Remove limitation to MySQL 5.7 and beyond
• Fix ProxySQL load users
• Fix flashing slave state in dashboard when slave in error
• Slimmer docker image (contributed by @erasys)

Changelog

• Improvements in failover election trying to give advantage to the most up to date slave
• Cancel failover when electable slave has issue
• Fix issues when replication manager was restarted with no current master
• Various dashboard fixes