feat(mcp): use a 10-minute sliding TTL for reviewer and merger PR leases (Closes #747)

Reviewer and merger PR leases minted a fixed 120-minute expiry (#407 AC6/AC7)
and derived staleness from two further activity bands: stale at 30 minutes,
reclaimable at 60. A session that died — daemon crash, transport flap, client
restart — therefore kept a PR blocked for an hour before anyone could reclaim
it, and two hours before manual cleanup was sanctioned. #718 records the
resulting deadlock: a merge stalled behind a reviewer lease that had stopped
being live long before it stopped being authoritative.

The flaw was using a long fixed expiry as a proxy for "the owner is probably
still alive" instead of making the owner continuously prove liveness.

Sliding window
--------------
`LEASE_TTL_MINUTES = 10` now governs acquisition, and every write of the lease
marker re-derives `expires_at` from the moment of the write, so each heartbeat
slides the window forward. An actively heartbeating session is never evicted
and has no maximum lifetime; a dead one releases its hold within one TTL.

`LEASE_RENEWAL_MINUTES` is named separately from the acquisition TTL. Renewal
previously had no seam at all: the heartbeat slid the expiry only as a side
effect of re-defaulting the acquisition constant, so the two durations could
not be reasoned about or tuned independently. `format_lease_body` now takes an
explicit `ttl_minutes`, and the heartbeat passes the renewal window rather than
relying on that default.

Merger leases acquire through the same lease-body formatter, so they inherit
the identical window by construction rather than by a parallel constant.

Removal of the reclaim tier
---------------------------
`classify_lease_freshness` no longer returns `reclaimable`. Beyond being an
extra waiting tier, that band is unreachable under a sliding TTL: a heartbeat
stamps `last_activity` and `expires_at` together, so a lease idle for a full
TTL is necessarily already expired. Expiry is now the only takeover gate.

No reclaim path is lost. `find_active_reviewer_lease` already ignores expired
markers, so an expired foreign lease never gated acquisition; and the
`foreign_expired` classification carries the same
`NEXT_ACTION_RELEASE_EXPIRED_LEASE` the retired `foreign_reclaimable` did. The
two updated tests in `test_reviewer_pr_lease.py` assert exactly that: the
classification label changes, the sanctioned next action does not.

`STALE_WARNING_MINUTES` drops to 5 — half the window — so the warning still
fires while the owner can heartbeat and recover.

Diagnostics
-----------
Adds `lease_seconds_remaining`, and the heartbeat tool now returns
`ttl_minutes`, `expires_at`, and `seconds_remaining`, so an operator can
distinguish "held and live" from "held and dying" instead of only seeing that
a lease exists. All additions are additive; no existing key changed.

Also removes `pr_work_lease.DEFAULT_REVIEWER_LEASE_TTL_MINUTES`, a duplicate of
the reviewer TTL with no readers anywhere in the tree, which could only drift.

Legacy markers minted under the old 120-minute TTL still parse and are judged
against their own recorded `expires_at`, so no lease is retroactively expired
by this change.

Full suite: 3425 passed, 6 skipped, 2 failed. Both failures
(`test_issue_702_review_findings_f1_f6`, `test_reconciler_supersession_close`)
reproduce identically on clean master b05075fd25 and are pre-existing.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01QxXHZ7rqXtLgTusngaWgKZ

gitea_mcp_server.py

@@ -12132,6 +12132,13 @@ def gitea_heartbeat_reviewer_pr_lease(
     verify_preflight_purity(remote, task="review_pr")
     h, o, r = _resolve(remote, host, org, repo)
     auth = _auth(h)
+    # Slide the sliding TTL forward (#747): the heartbeat is the liveness proof,
+    # so renewal is stated explicitly rather than inherited from the acquisition
+    # default.
+    beat_at = datetime.now(timezone.utc)
+    renewed_expiry = beat_at + timedelta(
+        minutes=reviewer_pr_lease.LEASE_RENEWAL_MINUTES
+    )
     body = reviewer_pr_lease.format_lease_body(
         repo=f"{o}/{r}",
         pr_number=pr_number,
@@ -12144,6 +12151,8 @@ def gitea_heartbeat_reviewer_pr_lease(
         candidate_head=candidate_head or session.get("candidate_head"),
         target_branch=session.get("target_branch") or "master",
         target_branch_sha=target_branch_sha or session.get("target_branch_sha"),
+        last_activity=beat_at,
+        ttl_minutes=reviewer_pr_lease.LEASE_RENEWAL_MINUTES,
     )
     comment_url = f"{repo_api_url(h, o, r)}/issues/{pr_number}/comments"
     with _audited(
@@ -12184,6 +12193,13 @@ def gitea_heartbeat_reviewer_pr_lease(
         "phase": phase,
         "comment_id": posted.get("id"),
         "session_lease": updated,
+        # Report the renewed window so an operator can tell "held and live"
+        # from "held and dying" (#747).
+        "ttl_minutes": reviewer_pr_lease.LEASE_RENEWAL_MINUTES,
+        "expires_at": renewed_expiry.replace(microsecond=0)
+        .isoformat()
+        .replace("+00:00", "Z"),
+        "seconds_remaining": reviewer_pr_lease.LEASE_RENEWAL_MINUTES * 60,
         "reasons": [],
     }
 

pr_work_lease.py

@@ -36,7 +36,8 @@ _TERMINAL_CONFLICT_FIX_PHASES = frozenset({"released", "blocked", "done"})
 _ACTIVE_CONFLICT_FIX_PHASES = frozenset({"claimed", "pushing", "pushed"})
 
 DEFAULT_CONFLICT_FIX_TTL_MINUTES = 120
-DEFAULT_REVIEWER_LEASE_TTL_MINUTES = 120
+# The reviewer/merger PR-lease TTL lives in reviewer_pr_lease.LEASE_TTL_MINUTES
+# (#747). A second copy here had no readers and could only drift out of sync.
 
 
 def _parse_timestamp(value: str | None) -> datetime | None:

reviewer_pr_lease.py

@@ -29,9 +29,19 @@ _ACTIVE_PHASES = frozenset({
     "adopted",
 })
 
-DEFAULT_LEASE_TTL_MINUTES = 120
-STALE_WARNING_MINUTES = 30
-RECLAIMABLE_MINUTES = 60
+# Reviewer and merger PR leases use a short *sliding* window (#747): a lease
+# expires 10 minutes after its last heartbeat, and every heartbeat slides the
+# expiry forward. An actively heartbeating session is never evicted, while a
+# dead session releases its hold in at most one TTL instead of the two hours
+# the previous fixed 120-minute expiry allowed.
+LEASE_TTL_MINUTES = 10
+# Renewal is named separately from acquisition so the slide amount is tunable
+# without silently re-defining how long a fresh lease lives.
+LEASE_RENEWAL_MINUTES = 10
+# Retained for callers that imported the pre-#747 name.
+DEFAULT_LEASE_TTL_MINUTES = LEASE_TTL_MINUTES
+# Warn at half the window, while the owner can still heartbeat and recover.
+STALE_WARNING_MINUTES = 5
 
 _SESSION_LEASE: dict[str, Any] | None = None
 
@@ -80,10 +90,19 @@ def format_lease_body(
     target_branch_sha: str | None,
     last_activity: datetime | None = None,
     expires_at: datetime | None = None,
+    ttl_minutes: int = LEASE_TTL_MINUTES,
     blocker: str = "none",
 ) -> str:
+    """Serialize a lease marker.
+
+    Every write of this marker — acquisition, heartbeat, adoption — re-derives
+    ``expires_at`` from the moment of the write, which is what makes the TTL
+    slide (#747). Callers renewing an existing lease pass
+    ``ttl_minutes=LEASE_RENEWAL_MINUTES``; an explicit ``expires_at`` still
+    wins so a lease can be minted with a deliberate window.
+    """
     now = last_activity or datetime.now(timezone.utc)
-    expires = expires_at or (now + timedelta(minutes=DEFAULT_LEASE_TTL_MINUTES))
+    expires = expires_at or (now + timedelta(minutes=ttl_minutes))
     last_text = now.astimezone(timezone.utc).replace(microsecond=0).isoformat().replace(
         "+00:00", "Z"
     )
@@ -169,8 +188,30 @@ def _minutes_since_activity(lease: dict, *, now: datetime) -> float | None:
     return (now - last).total_seconds() / 60.0
 
 
+def lease_seconds_remaining(lease: dict, *, now: datetime | None = None) -> int | None:
+    """Seconds until *lease* expires, clamped at 0; ``None`` if unparsable.
+
+    Lets diagnostics distinguish "held and live" from "held and dying" (#747)
+    rather than only reporting that a lease exists.
+    """
+    expires_at = _parse_timestamp(lease.get("expires_at"))
+    if not expires_at:
+        return None
+    now = now or datetime.now(timezone.utc)
+    return max(0, int((expires_at - now).total_seconds()))
+
+
 def classify_lease_freshness(lease: dict, *, now: datetime | None = None) -> str:
-    """Return active, stale_warning, reclaimable, expired, or terminal."""
+    """Return active, stale_warning, expired, or terminal.
+
+    Expiry is the only takeover gate (#747). The pre-#747 ``reclaimable`` band
+    sat between "stale" and "expired" and made a dead lease wait out a second
+    timer before anyone could reclaim it. Under a sliding TTL that band is also
+    unreachable: a heartbeat stamps ``last_activity`` and ``expires_at``
+    together, so a lease idle for a full TTL is already expired. Foreign
+    expired leases are handled by the ``foreign_expired`` classification, which
+    carries the same sanctioned release next-action the old tier did.
+    """
     now = now or datetime.now(timezone.utc)
     phase = (lease.get("phase") or "").strip().lower()
     if phase in _TERMINAL_PHASES:
@@ -180,8 +221,6 @@ def classify_lease_freshness(lease: dict, *, now: datetime | None = None) -> str
     minutes = _minutes_since_activity(lease, now=now)
     if minutes is None:
         return "active"
-    if minutes >= RECLAIMABLE_MINUTES:
-        return "reclaimable"
     if minutes >= STALE_WARNING_MINUTES:
         return "stale_warning"
     return "active"

tests/test_issue_747_sliding_lease_ttl.py

@@ -0,0 +1,210 @@
+"""Tests for the 10-minute sliding TTL on reviewer and merger PR leases (#747).
+
+The lease ledger previously minted a fixed 120-minute expiry and derived
+staleness from separate 30/60-minute activity bands. A dead session therefore
+held a PR for up to two hours. These tests pin the sliding-window contract:
+acquisition mints a 10-minute expiry, every heartbeat slides it forward, and an
+expired lease is immediately reclaimable with no intermediate waiting tier.
+"""
+
+import sys
+import unittest
+from datetime import datetime, timedelta, timezone
+
+sys.path.insert(0, str(__import__("pathlib").Path(__file__).resolve().parent.parent))
+
+import reviewer_pr_lease as leases
+
+
+def _body(
+    *,
+    session_id: str = "session-a",
+    pr_number: int = 747,
+    phase: str = "claimed",
+    last_activity: datetime | None = None,
+    expires_at: datetime | None = None,
+    ttl_minutes: int | None = None,
+) -> str:
+    kwargs = {}
+    if ttl_minutes is not None:
+        kwargs["ttl_minutes"] = ttl_minutes
+    return leases.format_lease_body(
+        repo="Scaled-Tech-Consulting/Gitea-Tools",
+        pr_number=pr_number,
+        issue_number=747,
+        reviewer_identity="rev1",
+        profile="prgs-reviewer",
+        session_id=session_id,
+        worktree="branches/review-pr747",
+        phase=phase,
+        candidate_head="a" * 40,
+        target_branch="master",
+        target_branch_sha="b" * 40,
+        last_activity=last_activity,
+        expires_at=expires_at,
+        **kwargs,
+    )
+
+
+def _comment(**kwargs) -> dict:
+    return {"id": 1, "body": _body(**kwargs), "user": {"login": "rev1"}}
+
+
+def _minutes_ago(minutes: int) -> datetime:
+    return datetime.now(timezone.utc) - timedelta(minutes=minutes)
+
+
+class TestSlidingTTLConstant(unittest.TestCase):
+    """AC6: one named constant per lease kind, no duplicated literals."""
+
+    def test_ttl_is_ten_minutes(self):
+        self.assertEqual(leases.LEASE_TTL_MINUTES, 10)
+
+    def test_renewal_window_is_separately_named(self):
+        self.assertEqual(leases.LEASE_RENEWAL_MINUTES, 10)
+
+
+class TestAcquisitionTTL(unittest.TestCase):
+    """AC1 / AC2: reviewer and merger acquisition both mint now + 10 minutes."""
+
+    def test_acquire_mints_ten_minute_expiry(self):
+        now = datetime(2026, 7, 18, 12, 0, 0, tzinfo=timezone.utc)
+        lease = leases.parse_lease_comment(_body(last_activity=now))
+        expires = leases._parse_timestamp(lease["expires_at"])
+        self.assertEqual(expires, now + timedelta(minutes=10))
+
+    def test_merger_acquisition_shares_the_same_window(self):
+        # Merger acquisition funnels through the same lease-body formatter, so
+        # the reviewer TTL is the merger TTL by construction.
+        now = datetime(2026, 7, 18, 12, 0, 0, tzinfo=timezone.utc)
+        lease = leases.parse_lease_comment(_body(phase="merging", last_activity=now))
+        expires = leases._parse_timestamp(lease["expires_at"])
+        self.assertEqual(expires, now + timedelta(minutes=10))
+
+
+class TestHeartbeatSlides(unittest.TestCase):
+    """AC3: a heartbeat slides expires_at to now + 10 minutes."""
+
+    def test_heartbeat_slides_expiry_forward(self):
+        acquired = datetime(2026, 7, 18, 12, 0, 0, tzinfo=timezone.utc)
+        beat = acquired + timedelta(minutes=7)
+        first = leases.parse_lease_comment(_body(last_activity=acquired))
+        renewed = leases.parse_lease_comment(_body(last_activity=beat))
+
+        first_expiry = leases._parse_timestamp(first["expires_at"])
+        renewed_expiry = leases._parse_timestamp(renewed["expires_at"])
+
+        self.assertEqual(renewed_expiry, beat + timedelta(minutes=10))
+        self.assertGreater(renewed_expiry, first_expiry)
+
+    def test_renewal_window_is_independently_tunable(self):
+        # The renewal amount must not be hardwired to the acquisition TTL;
+        # format_lease_body accepts an explicit window.
+        now = datetime(2026, 7, 18, 12, 0, 0, tzinfo=timezone.utc)
+        lease = leases.parse_lease_comment(_body(last_activity=now, ttl_minutes=3))
+        expires = leases._parse_timestamp(lease["expires_at"])
+        self.assertEqual(expires, now + timedelta(minutes=3))
+
+
+class TestFreshnessBands(unittest.TestCase):
+    """AC5: expiry is the only gate; no intermediate reclaim tier."""
+
+    def test_fresh_lease_is_active(self):
+        lease = leases.parse_lease_comment(_body(last_activity=_minutes_ago(1)))
+        self.assertEqual(leases.classify_lease_freshness(lease), "active")
+
+    def test_idle_past_half_ttl_warns_before_expiry(self):
+        lease = leases.parse_lease_comment(_body(last_activity=_minutes_ago(6)))
+        self.assertEqual(leases.classify_lease_freshness(lease), "stale_warning")
+
+    def test_lease_expires_after_ten_idle_minutes(self):
+        lease = leases.parse_lease_comment(_body(last_activity=_minutes_ago(11)))
+        self.assertEqual(leases.classify_lease_freshness(lease), "expired")
+
+    def test_no_separate_reclaimable_tier_remains(self):
+        # The old 60-minute reclaim band sat between "stale" and "expired" and
+        # blocked acquisition. Under a sliding TTL an idle lease is already
+        # expired, so the tier must not reappear at any idle duration.
+        for minutes in (11, 30, 65, 121, 600):
+            lease = leases.parse_lease_comment(_body(last_activity=_minutes_ago(minutes)))
+            self.assertEqual(
+                leases.classify_lease_freshness(lease),
+                "expired",
+                f"idle {minutes}m should be expired, not a waiting tier",
+            )
+
+
+class TestExpiredLeaseIsImmediatelyReclaimable(unittest.TestCase):
+    """AC5: another session takes over an expired lease with no extra wait."""
+
+    def setUp(self):
+        leases.clear_session_lease()
+
+    def _acquire_against(self, comments: list[dict]) -> dict:
+        return leases.assess_acquire_lease(
+            comments,
+            pr_number=747,
+            reviewer_identity="rev2",
+            profile="prgs-reviewer",
+            session_id="session-b",
+            repo="Scaled-Tech-Consulting/Gitea-Tools",
+            issue_number=747,
+            worktree="branches/review-pr747-b",
+            candidate_head="c" * 40,
+            target_branch="master",
+            target_branch_sha="d" * 40,
+        )
+
+    def test_expired_foreign_lease_does_not_block_acquisition(self):
+        comments = [_comment(session_id="dead-session", last_activity=_minutes_ago(11))]
+        result = self._acquire_against(comments)
+        self.assertTrue(result["acquire_allowed"], result["reasons"])
+
+    def test_live_foreign_lease_still_blocks_acquisition(self):
+        comments = [_comment(session_id="live-session", last_activity=_minutes_ago(2))]
+        result = self._acquire_against(comments)
+        self.assertFalse(result["acquire_allowed"])
+        self.assertTrue(any("already has active" in r for r in result["reasons"]))
+
+
+class TestRemainingTimeReporting(unittest.TestCase):
+    """AC7: diagnostics can distinguish 'held and live' from 'held and dying'."""
+
+    def test_seconds_remaining_on_live_lease(self):
+        now = datetime(2026, 7, 18, 12, 0, 0, tzinfo=timezone.utc)
+        lease = leases.parse_lease_comment(_body(last_activity=now))
+        remaining = leases.lease_seconds_remaining(lease, now=now + timedelta(minutes=4))
+        self.assertEqual(remaining, 360)
+
+    def test_seconds_remaining_is_zero_when_expired(self):
+        now = datetime(2026, 7, 18, 12, 0, 0, tzinfo=timezone.utc)
+        lease = leases.parse_lease_comment(_body(last_activity=now))
+        remaining = leases.lease_seconds_remaining(lease, now=now + timedelta(minutes=30))
+        self.assertEqual(remaining, 0)
+
+    def test_seconds_remaining_is_none_without_parsable_expiry(self):
+        self.assertIsNone(leases.lease_seconds_remaining({"expires_at": "not-a-time"}))
+
+
+class TestLegacyLeaseRows(unittest.TestCase):
+    """AC9: leases minted under the old 120-minute TTL still evaluate."""
+
+    def test_legacy_two_hour_expiry_is_honoured_until_it_passes(self):
+        now = datetime.now(timezone.utc)
+        legacy = leases.parse_lease_comment(
+            _body(last_activity=now - timedelta(minutes=90), expires_at=now + timedelta(minutes=30))
+        )
+        # Still inside its originally minted window: not expired, but idle long
+        # enough to warn.
+        self.assertEqual(leases.classify_lease_freshness(legacy), "stale_warning")
+
+    def test_legacy_row_past_its_own_expiry_is_expired(self):
+        now = datetime.now(timezone.utc)
+        legacy = leases.parse_lease_comment(
+            _body(last_activity=now - timedelta(minutes=180), expires_at=now - timedelta(minutes=60))
+        )
+        self.assertEqual(leases.classify_lease_freshness(legacy), "expired")
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_reviewer_pr_lease.py

@@ -79,22 +79,26 @@ class TestReviewerLeaseAcquire(unittest.TestCase):
 
 
 class TestReviewerLeaseFreshness(unittest.TestCase):
-    def test_stale_warning_after_30_minutes(self):
+    def test_stale_warning_at_half_the_sliding_window(self):
+        # #747 warns at half the 10-minute window, while the owner can still
+        # heartbeat and keep the lease.
         lease = leases.parse_lease_comment(
-            _lease_comment(382, "session-a", minutes_ago=35)["body"]
+            _lease_comment(382, "session-a", minutes_ago=6)["body"]
         )
         self.assertEqual(
             leases.classify_lease_freshness(lease),
             "stale_warning",
         )
 
-    def test_reclaimable_after_60_minutes(self):
+    def test_expired_once_the_sliding_window_lapses(self):
+        # Pre-#747 a 65-minute-idle lease was "reclaimable" and had to wait out
+        # a second timer. It is now simply expired and immediately takeable.
         lease = leases.parse_lease_comment(
             _lease_comment(382, "session-a", minutes_ago=65)["body"]
         )
         self.assertEqual(
             leases.classify_lease_freshness(lease),
-            "reclaimable",
+            "expired",
         )
 
 
@@ -281,7 +285,10 @@ class TestReviewerLeaseHandoffDiagnose(unittest.TestCase):
         self.assertEqual(result["active_lease"]["comment_id"], 8647)
         self.assertFalse(result["mutation_allowed"])
 
-    def test_foreign_reclaimable_release_expired(self):
+    def test_foreign_expired_release_expired(self):
+        # Pre-#747 this classified as "foreign_reclaimable" after the 60-minute
+        # activity band. Under the sliding TTL the lease is simply expired, and
+        # the sanctioned next action is unchanged.
         reclaim = _lease_comment(
             592, "foreign-old", phase="claimed", minutes_ago=65
         )
@@ -293,7 +300,7 @@ class TestReviewerLeaseHandoffDiagnose(unittest.TestCase):
             current_reviewer_identity="sysadmin",
             proposed_worktree="branches/review-pr-592",
         )
-        self.assertEqual(result["classification"], "foreign_reclaimable")
+        self.assertEqual(result["classification"], "foreign_expired")
         self.assertEqual(
             result["next_action"], leases.NEXT_ACTION_RELEASE_EXPIRED_LEASE
         )