Expected Behavior

The problem was reported by one of my customers as he expects that removing an endpoint and zone from the zones.conf after everything in the Director was moved to another one and then running the Kickstart does simply remove the objects also from the Director.

Current Behavior

It seems like there is an issue with the order on removal as the removal of the zone fails with "Failed to remove External Zone '...', it's eventually still in use", but the endpoint is removed successfully. Additional Kickstart runs do not even try to remove the zone and as external zone it could not be deleted manually, so only a delete in the database could remove the object.

Possible Solution

Perhaps it is really only the order on removal, but could also be much more complicated.

Steps to Reproduce (for bugs)

Before this all objects from the zone were moved to another one, the host object representing the satellite was also deleted and the configuration was deployed, so assume the zone is nowhere in use at the time.

1. Zone and endpoint was removed from zones.conf of both masters
2. Reload on both masters was executed successfully
3. Kickstart was run but failed with "Failed to remove External Zone '...', it's eventually still in use"
4. Deleting the zone from the databases was the only way to remove it

Your Environment

• Director version (System - About): 1.8.1
• Icinga Web 2 version and modules (System - About): 2.9.5
• Icinga 2 version (icinga2 --version): 2.13.1