From cbf6b02905711328ebb8b52db5065468a54cda62 Mon Sep 17 00:00:00 2001
From: hector <42570491+majinghe@users.noreply.github.com>
Date: Sun, 20 Sep 2026 11:27:56 +0800
Subject: [PATCH] ci(chain): fall back to auto-testing main HEAD when the
 pinned harness goes stale (#8026)

The functional chain pins the test harness to the revision recorded in
.config/functional-script-revision.txt, which is only refreshed by
release syncs - it went stale on Sep 16 and the nightly chain has been
replaying a Sep-17-era harness ever since. Last night that meant:
the performance lane died on a dpkg conffile prompt (fix merged as
#88 but never reached the chain), the FT lane died on 'unknown option:
--results-file' (harness/workflow version mismatch), and the
STS-105 signing fix (#94) sat untested.

When the pinned revision's commit is older than 24 hours, fall back to
auto-testing main HEAD so the chain always runs the current harness.
Any staleness-probe failure keeps the pinned revision (fail-safe).
The revision file stays in place as the audit trail and the release
sync continues to manage it.
---
 scripts/resolve_functional_candidate.py | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/scripts/resolve_functional_candidate.py b/scripts/resolve_functional_candidate.py
index 08f9eb9e4..7acdcd863 100644
--- a/scripts/resolve_functional_candidate.py
+++ b/scripts/resolve_functional_candidate.py
@@ -5,6 +5,7 @@ from __future__ import annotations
 import hashlib
 import io
 import json
+from datetime import datetime, timezone
 import os
 from pathlib import Path
 import re
@@ -112,6 +113,25 @@ def prepare():
     candidate = resolve(run_id, attempt)
     revision = (ROOT / ".config/functional-script-revision.txt").read_text().strip()
     require(sha(revision), "private test script revision must be pinned")
+    # The revision file is refreshed only by release syncs and goes stale
+    # within days, so the nightly chain keeps replaying an old harness while
+    # fixes (conffile prompts, FT --results-file, tier retries, STS-105
+    # signing) sit untested on auto-testing main. When the pinned revision is
+    # older than a day, fall back to auto-testing main HEAD. Any probe
+    # failure keeps the pinned revision - the chain still runs, just on the
+    # old harness.
+    try:
+        head = api("repos/rustfs/auto-testing/git/ref/heads/main")["object"]["sha"]
+        pinned_date = api(f"repos/rustfs/auto-testing/commits/{revision}")["commit"]["committer"]["date"]
+        pinned_at = datetime.fromisoformat(pinned_date.replace("Z", "+00:00"))
+        age_hours = (datetime.now(timezone.utc) - pinned_at).total_seconds() / 3600
+        if age_hours > 24:
+            print(f"functional-script revision {revision[:8]} is {age_hours:.0f}h old; falling back to auto-testing main HEAD {head[:8]}")
+            revision = head
+        else:
+            print(f"functional-script revision {revision[:8]} is {age_hours:.0f}h old; keeping pin")
+    except Exception as error:
+        print(f"staleness probe failed ({error}); keeping pinned revision {revision[:8]}")
     chain = {"schema": 1, "run_id": int(os.environ["GITHUB_RUN_ID"]), "attempt": int(os.environ["GITHUB_RUN_ATTEMPT"]),
              "workflow_sha": os.environ["GITHUB_SHA"], "testing_sha": revision, "candidate": candidate}
     encoded = json.dumps(chain, sort_keys=True, separators=(",", ":"))