Hii @Soelf,
This might be helpful for your first question. Icinga Dependencies and Implicit Dependencies for Services on Host.
Second question: It’s really hard to just tell you what the problem could be without letting us see your notification apply rule config. So, sharing this here would be quite helpful. But if I can guess, it somehow matches this completely unrelated host in this apply rule.
Third question: When you have configured your notification templates correctly, the following doc snipet should answer your question. Notifications.
Note
Only users who have been notified of a problem before (
Warning
,Critical
,Unknown
states for services,Down
for hosts) will receiveRecovery
notifications.