All of them are selfwritten checks, which are defined via the director and distributed via the director-global zone.
The only lead I have at the moment is the following message in the logs: Ignoring config update from endpoint 'icinga.xxxx.tld' for unknown zone 'director-global'.
Do you by chance use the Director’s Infrastructure section, where the director-global zone is configured in there, and the agents do not have this zone configured in their zones.conf file? Instead the global zone is synced via cluster config sync, and somehow the chicken egg problem worked before?
There are several fixes from the cluster config sync stages which make this handling more robust, thus unveiling errors in configuration made in the past.
this was exactly the problem removed the zone from the director infrastructure and added it to the zones.conf on all servers. Had to cleanup old configuration deployed from the director, but it all works now