fix(monitoring): bound Titan test metric labels

ci/scripts/publish_test_metrics.py

@@ -3,6 +3,7 @@
 
 from __future__ import annotations
 
+import hashlib
 import json
 import os
 from glob import glob
@@ -27,6 +28,8 @@ _infer_workspace_coverage_percent = _quality_helpers._infer_workspace_coverage_p
 _load_optional_json = _quality_helpers._load_optional_json
 _normalize_result_status = _quality_helpers._normalize_result_status
 
+TEST_CASE_LABEL_MAX_BYTES = 240
+
 
 def _escape_label(value: str) -> str:
     """Escape a Prometheus label value without changing its content."""
@@ -39,6 +42,17 @@ def _label_str(labels: dict[str, str]) -> str:
     return "{" + ",".join(parts) + "}" if parts else ""
 
 
+def _bounded_test_name(value: str) -> str:
+    """Keep test labels readable, unique, and safely below scraper limits."""
+    encoded = value.encode("utf-8")
+    if len(encoded) <= TEST_CASE_LABEL_MAX_BYTES:
+        return value
+    digest = hashlib.sha256(encoded).hexdigest()[:16]
+    suffix = f"...[sha256:{digest}]"
+    prefix = encoded[: TEST_CASE_LABEL_MAX_BYTES - len(suffix)]
+    return prefix.decode("utf-8", errors="ignore") + suffix
+
+
 def _read_text(url: str) -> str:
     """Fetch a plain-text response body from the given URL."""
     with urllib.request.urlopen(url, timeout=10) as response:
@@ -121,7 +135,7 @@ def _collect_junit_cases(pattern: str) -> list[tuple[str, str]]:
                     status = "failed"
                 elif test_case.find("skipped") is not None:
                     status = "skipped"
-                cases.append((full_name, status))
+                cases.append((_bounded_test_name(full_name), status))
     return cases
 
 
@@ -246,7 +260,7 @@ def _build_payload(
         for test_name, test_status in test_cases:
             labels = {
                 **test_case_base_labels,
-                "test": test_name,
+                "test": _bounded_test_name(test_name),
                 "status": test_status,
             }
             lines.append(

scripts/render/dashboards_render_atlas.py

@@ -1632,7 +1632,7 @@ def testing_case_variable():
         "label": "Test Case",
         "type": "query",
         "query": (
-            "query_result(topk(75, count by (test) (max_over_time("
+            "query_result(bottomk(75, min by (test) (min_over_time("
             f'platform_quality:test_case_health_rate:percent_1h{{suite=~"${{suite:regex}}",branch!="",'
             f'branch=~"${{branch:regex}}",test!="",test!="__no_test_cases__",'
             f'category!~"{PLATFORM_TEST_SUPPORT_CATEGORY_REGEX}"}}'

scripts/tests/test_dashboards_render_jobs.py

@@ -46,6 +46,8 @@ def test_jobs_dashboard_separates_current_gate_health_from_reliability():
 
     test_variable = variables_by_name["test"]
     assert test_variable["query"].startswith("query_result(")
+    assert "bottomk(75" in test_variable["query"]
+    assert "min_over_time" in test_variable["query"]
     assert "platform_quality:test_case_health_rate:percent_1h" in test_variable["query"]
     assert "platform_quality_gate_test_case_result" not in test_variable["query"]
     assert test_variable["regex"] == '/test="([^"]+)"/'

services/monitoring/dashboards/atlas-testing.json

@@ -5044,7 +5044,7 @@
         "name": "test",
         "label": "Test Case",
         "type": "query",
-        "query": "query_result(topk(75, count by (test) (max_over_time(platform_quality:test_case_health_rate:percent_1h{suite=~\"${suite:regex}\",branch!=\"\",branch=~\"${branch:regex}\",test!=\"\",test!=\"__no_test_cases__\",category!~\"fixtures|golden|helpers\"}[24h:1h]))))",
+        "query": "query_result(bottomk(75, min by (test) (min_over_time(platform_quality:test_case_health_rate:percent_1h{suite=~\"${suite:regex}\",branch!=\"\",branch=~\"${branch:regex}\",test!=\"\",test!=\"__no_test_cases__\",category!~\"fixtures|golden|helpers\"}[24h:1h]))))",
         "regex": "/test=\"([^\"]+)\"/",
         "current": {
           "text": "All",

services/monitoring/grafana-dashboard-testing.yaml

@@ -5053,7 +5053,7 @@ data:
             "name": "test",
             "label": "Test Case",
             "type": "query",
-            "query": "query_result(topk(75, count by (test) (max_over_time(platform_quality:test_case_health_rate:percent_1h{suite=~\"${suite:regex}\",branch!=\"\",branch=~\"${branch:regex}\",test!=\"\",test!=\"__no_test_cases__\",category!~\"fixtures|golden|helpers\"}[24h:1h]))))",
+            "query": "query_result(bottomk(75, min by (test) (min_over_time(platform_quality:test_case_health_rate:percent_1h{suite=~\"${suite:regex}\",branch!=\"\",branch=~\"${branch:regex}\",test!=\"\",test!=\"__no_test_cases__\",category!~\"fixtures|golden|helpers\"}[24h:1h]))))",
             "regex": "/test=\"([^\"]+)\"/",
             "current": {
               "text": "All",

testing/tests/test_hermes_deadline_http.py

@@ -247,6 +247,16 @@ def test_exchange_returns_result_and_enforces_response_limit():
         (b'{"ok":true,"status":"200","content_type":"a","body":""}', 0, "invalid evidence"),
         (b'{"ok":true,"status":200,"content_type":"a","body":"!!"}', 0, "invalid evidence"),
     ],
+    ids=[
+        "empty-error",
+        "oversized-output",
+        "invalid-json",
+        "false-result",
+        "array-result",
+        "missing-fields",
+        "empty-body",
+        "invalid-body",
+    ],
 )
 def test_exchange_rejects_invalid_helper_evidence(output, returncode, match):
     module = _load("deadline_http")

testing/tests/test_hermes_gitea_branch_coverage.py

@@ -141,6 +141,7 @@ def test_created_at_requires_short_timezone_aware_iso(value):
 @pytest.mark.parametrize(
     "value",
     [b"x" * (1024 * 1024 + 1), b"{}", json.dumps([None]).encode()],
+    ids=["oversized-input", "object-input", "null-list-input"],
 )
 def test_evaluate_rejects_invalid_top_level_inputs(value):
     module = _module("branch_evaluate_top")

testing/tests/test_publish_test_metrics_paths.py

@@ -65,6 +65,30 @@ def test_build_payload_includes_explicit_test_case_series():
     )
 
 
+def test_test_case_labels_are_bounded_and_collision_safe():
+    first = "module.test_case[" + ("x" * (4 * 1024 * 1024)) + "]"
+    second = first[:-2] + "y]"
+
+    first_label = publish_test_metrics._bounded_test_name(first)
+    second_label = publish_test_metrics._bounded_test_name(second)
+    payload = publish_test_metrics._build_payload(
+        suite="titan_iac",
+        status="ok",
+        tests={"tests": 2, "failures": 0, "errors": 0, "skipped": 0},
+        test_cases=[(first, "passed"), (second, "passed")],
+        ok_count=1,
+        failed_count=0,
+        branch="main",
+        build_number="1",
+        jenkins_job="titan-iac",
+    )
+
+    assert len(first_label.encode("utf-8")) <= publish_test_metrics.TEST_CASE_LABEL_MAX_BYTES
+    assert first_label != second_label
+    assert "[sha256:" in first_label
+    assert max(len(line.encode("utf-8")) for line in payload.splitlines()) < 1024
+
+
 def test_main_uses_reported_coverage_and_loc_without_fallback(tmp_path: Path, monkeypatch, capsys):
     build_dir = tmp_path / "build"
     build_dir.mkdir()