I'm trying to put together a dashboard / alerts to check on the success rate of firmware upgrades for some embedded devices. I want to display the results with Grafana but I can't quite figure out the logic.
In short, the device will (through an HTTP RESTful service) write an ES document just before the firmware upgrade starts (device ID, timestamp, version number) and then write another document after it comes back online a few minutes later.
I want to find the number of devices which have not come back online within, say, five minutes.
How might I do that?