CI "The job has exceeded the maximum execution time of N minutes"

By Daniel Zoghalchali·Latchkey

The runner stopped the job because it ran longer than the N-minute limit you set (or the default). The job did not crash - it was cut off at the bound, so the fix is a faster job or a longer, still-bounded timeout.

What this error means

A job is cancelled with The job has exceeded the maximum execution time of N minutes, where N is your timeout-minutes. It was still doing work at the limit. A transient slowdown sometimes tips an otherwise-in-budget job over; other times the limit is simply too low.

The job has exceeded the maximum execution time of 45 minutes.
##[error]The operation was canceled.

Common causes

A transient slowdown exceeded an otherwise-adequate limit

A slow dependency fetch or a congested runner pushed a job that normally fits within N minutes just over the line. A retry usually lands back in budget.

The configured limit is too tight for the work

If the job consistently needs more than N minutes, it hits the limit every run. Either raise timeout-minutes or cut the runtime down.

How to fix it

Right-size the timeout

Pick an N with headroom over the real p95 runtime, not the average.

jobs:
  test:
    runs-on: ubuntu-latest
    timeout-minutes: 45   # headroom over the real worst-case

Reduce the runtime

1. Cache dependencies and build outputs to cut repeated work.
2. Shard tests/builds across parallel jobs.
3. Add per-step timeouts so one slow step fails fast rather than eating the whole budget.

How to prevent it

• Set timeout-minutes based on p95 runtime with headroom.
• Cache and parallelize to keep runtimes under the bound.
• Alert on duration trending toward the limit so you adjust before it fails.

Frequently asked questions

Related guides

References

• GitHub Actions documentation