infrastructure//timeout

A timeout is the maximum time a serverless function is allowed to run before the platform kills it. If the function has not responded by then, it dies, with no graceful shutdown and no retry. Timeouts exist because serverless platforms bill per execution time and share resources: letting one function run forever would starve others and rack up costs.

A timeout is the maximum time a serverless function is allowed to run before the platform kills it. If the function has not responded by then, it dies, with no graceful shutdown and no retry. Timeouts exist because serverless platforms bill per execution time and share resources: letting one function run forever would starve others and rack up costs.

Vercel Functions10 s (Hobby), 60 s (Pro), 300 s (Enterprise).

Cloudflare WorkersTied to CPU time, not wall time: 10 ms (free), 30 s (paid), 5 min at most.

AWS LambdaUp to 15 minutes.

The timeout is why heavy batch processing (scanning thousands of emails, for instance) cannot run in a single serverless function: it gets killed mid-work. Job queues like Inngest or Trigger.dev solve it by breaking the work into small chunks, each within the timeout, and Vercel Cron Jobs handle recurring scheduled work.