Worker exceeded CPU time limit
We’re encountering issues where Workers perform reliably for a while but eventually start failing with the error:
Worker exceeded CPU time limit. I’m trying to better understand how CPU time consumption is calculated. I’ve reviewed the documentation at Cloudflare Workers CPU Time Limits https://developers.cloudflare.com/workers/platform/limits/#cpu-time, but I still have a few questions:
1) If a DurableObject sets an alarm, does the alarm callback continue consuming the same CPU time budget, or does it receive a fresh, separate budget?
2) In the case where a Worker is being non-stop invoked via RPCs, does each invocation receive a seperate CPU time budget? Or does the Worker need to be inactive for a period before its CPU time budget resets? If a cooldown or reset period is required, what is the duration?1 Reply
@Walshy | DevPlat EE could you, please, take a look ^^