GitHub Actions step killed because timeout-minutes was exceeded

By Kaveh Alemi·Latchkey

A step or job ran longer than its timeout-minutes value, so the runner cancelled it. This is a configured limit, not infrastructure failure - the step either needs more time or is hanging.

What this error means

A step is cancelled mid-run with a message that the timeout was reached, even though it had not finished.

##[error]The action 'Run integration tests' has timed out after 10 minutes.
##[error]The operation was canceled.

Common causes

Limit lower than the real runtime

timeout-minutes was set tighter than the step actually needs, so a healthy run gets cut off.

Step is hanging

A process waits on input, a network call, or a service that never responds, so it never completes within the limit.

How to fix it

Right-size the timeout or fix the hang

1. Measure the step's normal duration and set timeout-minutes with headroom.
2. If it hangs, add internal timeouts to the hanging command (for example a test or curl timeout).
3. Re-run.

- name: Integration tests
  timeout-minutes: 20
  run: npm run test:integration -- --testTimeout=60000

How to prevent it

• Set timeout-minutes with headroom over observed runtimes.
• Add command-level timeouts so hangs fail fast with a clear cause.

Frequently asked questions

Related guides

References

• GitHub Actions documentation
• Workflow syntax for GitHub Actions