[muzimu217]

What this fixes

The Trusted extensions Electron E2E job can hang instead of failing. On #1751 it ran 30m18s and was cancelled by the job budget with no failing step to read — conclusion: cancelled, zero failed steps.

The cause is this runner's only timeout:

const timer = setTimeout(() => child.kill(), 60_000);

SIGTERM is a request. An Electron child still initialising its renderer can ignore it, and the close event this script awaits then never arrives — so the runner's own 60-second budget is never enforced and the step holds the job until GitHub cancels it.

Change

Escalate, and say which signal fired:

• SIGTERM at 60s, SIGKILL at 65s
• assert killedBy === null, so a probe that cannot start fails this step in about a minute with its output attached instead of burning the job budget

Verification

• Normal path unchanged: a real run on this branch prints baseline: Electron 43 net.fetch fails before HTTP then provider client certificate E2E passed.
• Timeout path terminates: with a stub child that ignores SIGTERM, the escalation force-kills it at ~804ms in an isolated harness, confirming the timeout no longer hangs.
• node --check clean; branch is one commit on current main (a6f4ba121), 1 file +23/−2.

Why this matters beyond one job

The same escalation exists on #1759's branch (its author hit the same hang) but that PR is still open, so main carries the hazard. Any Electron E2E step that hangs here costs 30 minutes of runner time and returns no diagnostic — this makes it a one-minute failure that names the signal.