Is execution timeout per request or per worker execution?
https://docs.runpod.io/serverless/endpoints/send-requests#--execution-policy
"Execution Timeout: Specifies the maximum duration that a job can run before it's automatically terminated."
The endpoint edit UI says - "Maximum amount of time in seconds a request can run for."
I read the first one as "max lifetime of a worker" - i.e. if it takes 5s to process a request and execution timeout is 60s, the worker will process 12 requests and die.
I read the first one as "if a request takes 60s, the worker will die, but as long as requests take <60s, the worker will run forever"
which one is it?
Send a request | RunPod Documentation
The method in which jobs are submitted and returned.
2 Replies
I would appreciate the answer. Thank you!
It is applied to a single request, not 12 requests
They both do the same thing. The endpoint config applies to all requests but you can override it on a per request basis, as the docs specify.
Its pretty clear in the docs, so I don't know why you are thinking it applies to multiple requests when it applies to 1 request at a time.