Anyone have a good automation / blueprint / alert / script / something to trigger klaxons if the recorder process dies? Last two days apparently all stats were just gone and I didn’t notice because I was away from home.
- Although it is an interesting task - a more important task is “minimize a possibility of Recorder death”.
- Can you see any error messages in Log at the moment when “recorder dies”? If yes - then you can create an automation triggered by event “sys_error_log” (if I am not mistaken), analyze a message - if it contains anything about “recorder dies” then do something.
1 Like
it was some massive query that was blocking the recorder process, no log files associated with it, sadly. So need something to look for metrics not being committed.
Erm, I meant the “main log” - does it have any errors related to recorder.