My App Engine app reports high frontend usage. For example:
Usage: 3.1h of 35h
Load: /user 2s
How is this computed?
i think the metric reckons active time serving reqs, not merely uptime. maybe its influenced by scale-ups or idle periods? have u seen if sudden load bumps cause irregular patterns? curious what you thnk about how it monitors burst activity over time
i think it only counts when your instance is actively handling reqs. if its idle, it doesnt add to frontend hours so its more akin to actual work time rather than pure uptime