Cloud PostgreSQL Health Check monitoring

[James_Eads]

What do others use,  Postgresql Telemetry pretty lacking in appd.   I ended up writing a few of my own custom metrics. 
I have the following :
Use data from last:5 minutes, wait time after violation 1 minute

Critical Criteria 

DB availability : single metric < 1,  trigger only when violation occurs 5 times within 5 minutes

Custom Metric: — created a query to determined blocked sessions then trigger off 
single metric < 1,  trigger only when violation occurs 5 times within 5 minutes

Warning Criteria

Custom metric for checking bloat
Custom metric for session counts based on baseline, as KPI for numbackends proved to be inaccurate
custom metric for long running queries as there isn’t any KPI for that. 

Anyone else do anything differently?

[Akshit_Grover]

@James.Eads 

For DB availability I'd suggest to use this "connection down" event rather then the availability metric which will also be triggered when the agent goes down!

https://docs.appdynamics.com/appd/24.x/latest/en/database-visibility/monitor-databases-and-database-...