ci(e2e): time-box the artifact download retry and the two uncapped legs (#15917)

Each leg downloads the shared build archive twice: once, and again from
the retry that only fires when the first attempt failed. Only the first
attempt carried timeout-minutes, so a retry that stalls held its runner
with no ceiling at all — up to GitHub's 360-minute default. Run
37119220504 (#13284) already showed both attempts lost inside one
window on a macOS shard while its sibling downloaded the same artifact.

Two limbs:

- Give the four 'Download build artifact (retry)' steps
  timeout-minutes: 10. The value is not new: the four first-attempt
  downloads in the same jobs already pin 10.
- Give e2e-test-macos and isolated-nightly a job-level
  timeout-minutes: 60. The two other artifact-consuming legs
  (e2e-test-linux, e2e-interactive-opentui) already use 60, and 60 is
  also the ceiling the shard-retry budget gate in run-e2e-tests.sh
  takes its 25-minute reserve out of. `build` keeps its own 30 — it
  produces the archive rather than consuming it.

web-shell-browser-regression has the same missing job ceiling but is
left alone: it is outside this finding's scope and does not consume the
build artifact.

The retry steps keep their absence of continue-on-error — the first
attempt's continue-on-error is what makes the retry reachable, and
copying it onto the retry would let a stalled download pass silently
into the unpack step. isolated-nightly's job-level continue-on-error is
untouched, so its 60-minute ceiling frees the runner without changing
the leg's conclusion.

Pinned in scripts/tests/e2e-workflow.test.js: both limbs, the retry's
visible failure, and a regression guard over the already-landed values
(build 30, the two consuming legs 60, the four first-attempt steps 10).

Refs #12132


Patrol-Run: qwen-issue-patrol/jmvuv88frpw

Co-authored-by: yiliang114 <[email protected]>
Co-authored-by: Qwen-Coder <[email protected]>

.github/workflows/e2e.yml

@@ -385,6 +385,7 @@ jobs:
       - name: 'Download build artifact (retry)'
         if: "${{ steps.download-build.outcome == 'failure' }}"
         uses: 'actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c' # v8.0.1
+        timeout-minutes: 10
         with:
           name: 'e2e-build'
           path: '${{ runner.temp }}/e2e-build'
@@ -465,6 +466,7 @@ jobs:
   e2e-test-macos:
     name: 'E2E Test - macOS - shard ${{ matrix.shard }}'
     runs-on: 'macos-latest'
+    timeout-minutes: 60
     needs:
       - 'build'
     # Skip on fork PRs (no secrets) — see e2e-test-linux above.
@@ -554,6 +556,7 @@ jobs:
       - name: 'Download build artifact (retry)'
         if: "${{ steps.download-build.outcome == 'failure' }}"
         uses: 'actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c' # v8.0.1
+        timeout-minutes: 10
         with:
           name: 'e2e-build'
           path: '${{ runner.temp }}/e2e-build'
@@ -671,6 +674,7 @@ jobs:
       - name: 'Download build artifact (retry)'
         if: "${{ steps.download-build.outcome == 'failure' }}"
         uses: 'actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c' # v8.0.1
+        timeout-minutes: 10
         with:
           name: 'e2e-build'
           path: '${{ runner.temp }}/e2e-build'
@@ -699,6 +703,7 @@ jobs:
   isolated-nightly:
     name: '${{ matrix.label }} (nightly)'
     runs-on: 'ubuntu-latest'
+    timeout-minutes: 60
     needs:
       - 'build'
     if: |-
@@ -785,6 +790,7 @@ jobs:
       - name: 'Download build artifact (retry)'
         if: "${{ steps.download-build.outcome == 'failure' }}"
         uses: 'actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c' # v8.0.1
+        timeout-minutes: 10
         with:
           name: 'e2e-build'
           path: '${{ runner.temp }}/e2e-build'

scripts/tests/e2e-workflow.test.js

@@ -1342,6 +1342,80 @@ describe('e2e workflow', () => {
     });
   });
 
+  describe('artifact download time-boxing', () => {
+    // #12132: each leg downloads the shared archive twice — once, and again
+    // from the retry that only fires when the first attempt failed. Only the
+    // first attempt was time-boxed, so a stalling retry held its runner with
+    // no ceiling at all, and two legs had no job-level ceiling either.
+    const legs = [
+      'e2e-test-linux',
+      'e2e-test-macos',
+      'e2e-interactive-opentui',
+      'isolated-nightly',
+    ];
+    const downloads = (leg) =>
+      yml.jobs[leg].steps.filter((step) =>
+        String(step.name).startsWith('Download build artifact'),
+      );
+
+    it('caps every artifact-consuming leg at the job level', () => {
+      // 60 is the ceiling e2e-test-linux and e2e-interactive-opentui already
+      // carried; the other two legs consume the same archive on the same
+      // runners, so they get that budget instead of a new number.
+      for (const leg of legs) {
+        expect(yml.jobs[leg]['timeout-minutes'], leg).toBe(60);
+      }
+    });
+
+    it('time-boxes the retry as tightly as the first attempt', () => {
+      // 10 is the value every first-attempt download already pins. Without it
+      // the retry — the step that exists precisely because the first attempt
+      // misbehaved — was the only unbounded step in the download leg.
+      for (const leg of legs) {
+        const steps = downloads(leg);
+        expect(
+          steps.map((step) => step.name),
+          leg,
+        ).toEqual([
+          'Download build artifact',
+          'Download build artifact (retry)',
+        ]);
+        for (const step of steps) {
+          expect(step['timeout-minutes'], `${leg}/${step.name}`).toBe(10);
+        }
+      }
+    });
+
+    it('keeps the retry step red when the retry itself fails', () => {
+      // The first attempt's continue-on-error is what makes the retry
+      // reachable. Copying it onto the retry would leave a stalled download
+      // invisible: the leg would go on to unpack a missing archive and report
+      // a failure with the wrong cause.
+      for (const leg of legs) {
+        const retry = downloads(leg)[1];
+        expect(retry.name, leg).toBe('Download build artifact (retry)');
+        expect(retry['continue-on-error']).toBeUndefined();
+      }
+    });
+
+    it('leaves the already-landed ceilings where they are', () => {
+      // Regression guard. `build` produces the archive on a 30-minute budget
+      // while the legs consume it on 60, and 60 is also the number the
+      // shard-retry budget gate in run-e2e-tests.sh takes its 25-minute
+      // reserve out of. A later pass that "unifies" these values would move
+      // either the producer's ceiling or the retry arithmetic with every
+      // other witness green.
+      expect(yml.jobs.build['timeout-minutes']).toBe(30);
+      expect(yml.jobs['e2e-test-linux']['timeout-minutes']).toBe(60);
+      expect(yml.jobs['e2e-interactive-opentui']['timeout-minutes']).toBe(60);
+      for (const leg of legs) {
+        const first = downloads(leg)[0];
+        expect(first['timeout-minutes'], leg).toBe(10);
+        expect(first['continue-on-error'], leg).toBe(true);
+      }
+    });
+  });
+
   describe('one build for every leg', () => {
     // Each leg used to build and bundle on its own runner — 4–8 minutes on a
     // hosted VM, 10–17 on a busy pool host, eleven times per run. The `build`