Ceph » RADOS

Hm. what do yo expect? Upgrade scenarios can become complicated with more than two versions running at the same time while doing an upgrade. What do you have in mind?

I have completed a function called check_daemon_version located in src/mon/Monitor.cc This function goes through mon_metadata checking the version of each daemon against the version of the first daemon, if there are any that do not match their rank is added to a map function and returned along with the value true, once returned the map is cycled through in src/mon/HealthMonitor.cc file function get_health_status() where the rank is used to get the name of each daemon and return them. Their is an if statement inside HealthMintor.cc that ensures even if check_daemon_version returns true the error message will only occur if a specific amount of time has passed and check_daemon_version returns true again.
If you wish to take a look at the code you can check the current pull request located here: https://github.com/ceph/ceph/pull/35876

Delay 7 days by default with a config value before health warning/error when detected.