How do you put alerts on your node in case it’s messing up
For participation nodes, look at app.metrika.co that does some monitoring for you.
You can also parse the logs to count manually the votes and block proposals and compare to what you should be doing.
I also like to have some system comparing the last block of the node (endpoint /v2/status) with a public API service regularly via a cron job/lambda function (Ecosystem Tools & Projects | Algorand Developer Portal). So you can check for sync issues.
Finally, you should do normal monitoring of your node: CPU use, RAM use, disk use, network use, and have alerts for all of those when you go above some thresholds.
2 Likes
Is there calculator for that i.e to know what i should be doing? for the ram, cpu, disk and network what’s a good threshold to look at.