I looked at the documentation of sentry (have not installed it). In my opinion, it uses all the parts which a modern nagios replacement should have: task-queue (celery, rabbit-mq), ...
I like to use structured logging in the future. And here sentry seems to be a good choice
The missing part is just a sheduler which calls check scripts and then sends the result like a structured logging information. We already have plenty of
check-scripts. It would be easy to return their output into structured logging events.
Since our current server monitoring solution is quite simple, using sentry could be a solution for us. If we go this way, I will write here again.
Regards, Thomas Güttler