I wanted to have some sort of failover for my XCP 1.6 virtual environment prior to going to that full time with backups and restore procedures in place plus some failover capability using HA-Lizard. I got a great indication I’m heading in the correct direction when a couple of hours ago I started shutting down the VM’s on my XCP system and they kept restarting.
A big duh I had implemented HA-Lizard Saturday evening verified settings and was enabled on my shared pool on XCP and all VM’s residing on those servers within those pools. So as I shutdown each VM gracefully from the OS I noticed that each time it restarted.
I went on the CLI of my Master XCP server in the pool and did the following to finally shutdown first the HA-Lizard environment and then the XCP environment until later when I make more additions and have time to work on it:
- Go to the XCP Master Server Console
- #service ha-lizard status –>> verified HA-Lizard running
- #service ha-lizard stop –>> Stopped HA-Lizard service but HA-Lizard watchdog service restarted HA-Lizard service
- #service ha-lizard status –>> verified HA-Lizard HA service running after started by HA-Lizard watchdog service
- #service ha-lizard -w stop –>> Stops HA-Lizard watchdog service and HA-Lizard HA service
- #service ha-lizard -w status –>> Verify HA-Lizard watchdog service stopped
- #service ha-lizard status –>> Verify HA-Lizard HA service stopped
- Shutdown each VM from OS and they now stayed shutdown.
- Placed XCP servers in maintenance mode and then shut them down.
I also noticed that before I shutdown the HA-Lizard services if I went to place the XCP host master server or slave server in maintenance mode I was notified it would migrate my VM guest OS’! How sweet is that!
This HA-Lizard XCP combination may be just what I have been looking for a truly Open Source virtual HA environment !
