Quarantine transport-invalid oracle trial

docs/opprof/oracle-gap-protocol.md

@@ -1,11 +1,29 @@
 # Static-policy oracle-gap protocol
 
-Status: **FROZEN WITH A-OG-1 AMENDMENT**.
+Status: **FROZEN WITH A-OG-1 AND A-OG-2 AMENDMENTS**.
 
 Date frozen: 2026-07-13 (Asia/Singapore). Existing Phase-3 measurements were
 inspected only to choose the workload pair and rate brackets. They are
 exploratory calibration data, not primary observations in this protocol.
 
+### A-OG-2 — retry one transport-invalid attempt (after 49 scored trials)
+
+The first attempt at `P06-C01-r2.2-rep0` produced one local
+`ClientOSError` 3.27 ms after admission, with HTTP status 0, no first token,
+and no output. The other 399 requests completed successfully, every successful
+request produced exactly 512 tokens, the server stayed healthy, and all other
+client invariants passed. The controller rejected the attempt before creating
+a `score.json`; no TTFT/TPOT result from this attempt was used to choose the
+recovery rule.
+
+This is a measurement-transport failure, not an observed server SLO outcome.
+The entire attempt is content-hashed and moved under `invalid-attempts/`, then
+the identical logical trial (phase, config, rate, repetition, derived manifest,
+seeds, timeline, and SLO) is run once on a fresh C01 server. All 49 existing
+scores remain immutable. A second client transport failure in the retry is a
+stop condition and makes the experiment inconclusive; it must not be retried
+again. No grid, policy, threshold, order, or oracle calculation changes.
+
 ### A-OG-1 — extend the P06 upper bracket (after trial 33)
 
 The controller stopped as registered after C00/P06 remained SLO-feasible at

scripts/oracle_gap/run_frontier.py

@@ -20,7 +20,11 @@ from analyze import CONFIGS, PHASES, atomic_json, score_trial, summarize_trials
 
 
 SCHEMA = 1
-AMENDMENT = "A-OG-1"
+AMENDMENT = "A-OG-2"
+AMENDMENT_REASON = (
+    "retry one quarantined P06-C01-r2.2-rep0 transport-invalid attempt "
+    "without changing its logical trial"
+)
 REMOTE_ROOT = Path("/home/admin/cpfs/wjh/oracle-gap-20260713")
 RUN_ROOT = REMOTE_ROOT / "runs"
 STATE = RUN_ROOT / "controller-state.json"
@@ -235,6 +239,7 @@ def fingerprint() -> dict[str, Any]:
 
 def resume_compatible(old: dict[str, Any], current: dict[str, Any]) -> bool:
     immutable = (
+        "analyzer_sha256",
         "p5_client_sha256",
         "p3_client_sha256",
         "vllm_commit",
@@ -245,7 +250,21 @@ def resume_compatible(old: dict[str, Any], current: dict[str, Any]) -> bool:
         "config_details",
         "base_rates",
     )
-    return all(old.get(key) == current.get(key) for key in immutable)
+    if not all(old.get(key) == current.get(key) for key in immutable):
+        return False
+    old_extensions = old.get("up_extensions")
+    current_extensions = current.get("up_extensions")
+    if old_extensions == current_extensions:
+        return True
+    # The only earlier compatible transition was A-OG-1, which appended the
+    # five registered P06 anchors after the original 2.3-rps list.
+    if not isinstance(old_extensions, dict) or not isinstance(current_extensions, dict):
+        return False
+    return (
+        old_extensions.get("P01") == current_extensions.get("P01")
+        and list(old_extensions.get("P06", [])) + [2.4, 2.5, 2.6, 2.8, 3.0]
+        == current_extensions.get("P06")
+    )
 
 
 def ensure_provenance(current: dict[str, Any]) -> None:
@@ -650,10 +669,7 @@ def execute(resume: bool) -> None:
             {
                 "id": AMENDMENT,
                 "accepted_at": time.time(),
-                "reason": (
-                    "C00/P06 feasible through the original 2.3-rps "
-                    "upper extension"
-                ),
+                "reason": AMENDMENT_REASON,
                 "completed_trials_before": len(
                     state.get("completed_trials", [])
                 ),

tests/test_oracle_gap.py

@@ -9,7 +9,7 @@ ORACLE_GAP = Path(__file__).resolve().parents[1] / "scripts/oracle_gap"
 sys.path.insert(0, str(ORACLE_GAP))
 
 from analyze import score_trial, summarize_trials  # noqa: E402
-from run_frontier import UP_EXTENSIONS, resume_compatible  # noqa: E402
+from run_frontier import AMENDMENT, UP_EXTENSIONS, resume_compatible  # noqa: E402
 
 
 def _request(
@@ -153,14 +153,19 @@ def test_a_og_1_extends_only_mutable_resume_fields() -> None:
         "manifests": {"P01": "a", "P06": "b"},
         "runtime": "runtime",
         "driver": "driver",
+        "analyzer_sha256": "analyzer",
         "config_details": {"C00": {}},
         "base_rates": {"P01": [26], "P06": [1.4]},
+        "up_extensions": {"P01": [38.0], "P06": [2.1, 2.2, 2.3]},
     }
     amended = {
         **immutable,
         "controller_sha256": "new",
         "repo_commit": "new",
-        "up_extensions": {"P06": [2.4]},
+        "up_extensions": {
+            "P01": [38.0],
+            "P06": [2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.8, 3.0],
+        },
     }
 
     assert UP_EXTENSIONS["P06"][-5:] == (2.4, 2.5, 2.6, 2.8, 3.0)
@@ -168,3 +173,26 @@ def test_a_og_1_extends_only_mutable_resume_fields() -> None:
     assert not resume_compatible(
         immutable, {**amended, "manifests": {"P01": "changed"}}
     )
+
+
+def test_a_og_2_requires_the_amended_grid_to_stay_fixed() -> None:
+    old = {
+        "analyzer_sha256": "analyzer",
+        "p5_client_sha256": "p5",
+        "p3_client_sha256": "p3",
+        "vllm_commit": "vllm",
+        "model": "model",
+        "manifests": {"P01": "a", "P06": "b"},
+        "runtime": "runtime",
+        "driver": "driver",
+        "config_details": {"C00": {}},
+        "base_rates": {"P01": [26], "P06": [1.4]},
+        "up_extensions": {"P01": [38.0], "P06": [2.1, 2.2, 2.3, 2.4]},
+    }
+
+    assert AMENDMENT == "A-OG-2"
+    assert resume_compatible(old, {**old, "controller_sha256": "new"})
+    assert not resume_compatible(
+        old,
+        {**old, "up_extensions": {"P01": [38.0], "P06": [2.1, 2.2, 2.3, 2.5]}},
+    )