PowerShell long-running job timeout in CI

A PowerShell background job or a long external command ran past its time budget and the runner cancelled the step. Sometimes the work is genuinely slow; sometimes it stalled on a transient hang.

What this error means

The step is killed with a timeout/cancellation message after running long, often during Wait-Job, a network wait, or a heavy build. Intermittent when the cause is a transient stall.

Wait-Job : The job has not completed within the timeout period.
##[error]The operation was canceled.
##[error]The job running on runner ... has exceeded the maximum execution time.

Common causes

No explicit timeout, so the default kills it

Without a bounded Wait-Job timeout or a step timeout-minutes, a stalled job runs until the job-level limit, wasting time before failing.

A transient stall in a network or external dependency

A hung download, registry call, or external service can stall the job until the runner cancels it, even though a retry would succeed quickly.

How to fix it

Bound the wait and fail fast

Give Wait-Job a timeout so a stall fails quickly and can be retried instead of burning the whole budget.

$job = Start-Job { ./long-task.ps1 }
if (-not (Wait-Job $job -Timeout 600)) {
  Stop-Job $job; throw 'job timed out'
}
Receive-Job $job

Set a step-level timeout

Cap the step so a hang does not consume the whole job limit.

- name: Long task
  timeout-minutes: 15
  shell: pwsh
  run: ./long-task.ps1

How to prevent it

• Bound waits with explicit timeouts and set step timeout-minutes. Transient stalls are common; on Latchkey managed runners self-healing auto-retries transient timeout-class failures so a one-off stall does not fail the job. Windows minutes bill at roughly 2x Linux, so move long, slow steps to Linux where the work allows.

Related guides