Hoping the experts can help me with a simple problem.
I have a simple flow that I'm using to check for the presence of "heartbeat" events with the em_event table. The flow runs every 30 minutes, looking for a qualifying event within the 60 minute look-back period defined in the flow. All of this works flawlessly! (Thanks to a similar post in this forum on the topic).
The issue I've run into is the source system that sends the heartbeat notifications doesn't send them "on the hour" reliably. Their arrival tends to slip every hour by a 1/2 minute or more. Over a period of a few days, my flow executes before the heartbeat is received. This results in a false positive & unnecessary critical incident.
My thought process was to simply change the time at which the flow executes. I thought I could simply deactivate the flow, wait until the desired new execution time, activate the flow and I'd have a new execution time. Unfortunately, I've tried that approach twice and the execution time remains unchanged.
I do know that I could increase the look-back period to reduce false positives. However, if I do so I'm concerned that I could also extend the duration of an outage by not identifying the failure in a timely manner.
Appreciate any assistance you can provide.
Here's the simple trigger definition and below is the full flow.
