[pull] main from SigNoz:main

pkg/prometheus/engine.go

@@ -2,6 +2,7 @@ package prometheus
 
 import (
 	"log/slog"
+	"time"
 
 	"github.com/prometheus/prometheus/promql"
 )
@@ -23,5 +24,11 @@ func NewEngine(logger *slog.Logger, cfg Config) *Engine {
 		Timeout:            cfg.Timeout,
 		ActiveQueryTracker: activeQueryTracker,
 		LookbackDelta:      cfg.LookbackDelta,
+		// The engine calls this for subqueries that do not set a step, such as
+		// `metric[5m:]`, and segfaults if it is nil. 1m matches the default
+		// global evaluation_interval that Prometheus wires here.
+		NoStepSubqueryIntervalFn: func(int64) int64 {
+			return time.Minute.Milliseconds()
+		},
 	})
 }

pkg/prometheus/engine_test.go

@@ -0,0 +1,33 @@
+package prometheus
+
+import (
+	"context"
+	"log/slog"
+	"testing"
+	"time"
+
+	"github.com/prometheus/prometheus/storage"
+	"github.com/stretchr/testify/require"
+)
+
+func TestNoStepSubqueryDoesNotPanic(t *testing.T) {
+	engine := NewEngine(slog.New(slog.DiscardHandler), Config{Timeout: time.Minute})
+	queryable := storage.QueryableFunc(func(int64, int64) (storage.Querier, error) {
+		return storage.NoopQuerier(), nil
+	})
+
+	qry, err := engine.NewRangeQuery(
+		context.Background(),
+		queryable,
+		nil,
+		"max_over_time(some_metric[5m:])",
+		time.Now().Add(-time.Hour),
+		time.Now(),
+		time.Minute,
+	)
+	require.NoError(t, err)
+	defer qry.Close()
+
+	res := qry.Exec(context.Background())
+	require.NoError(t, res.Err)
+}

tests/integration/testdata/alerts/test_scenarios/promql_subquery_no_step/alert_data.jsonl

@@ -0,0 +1,5 @@
+{"metric_name":"cpu_percent_promql_subquery_no_step","labels":{"host":"server-01","cpu":"cpu0"},"timestamp":"2026-01-29T10:01:00+00:00","value":15,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false,"flags":0,"description":"","unit":"","env":"default","resource_attrs":{},"scope_attrs":{}}
+{"metric_name":"cpu_percent_promql_subquery_no_step","labels":{"host":"server-01","cpu":"cpu0"},"timestamp":"2026-01-29T10:02:00+00:00","value":15,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false,"flags":0,"description":"","unit":"","env":"default","resource_attrs":{},"scope_attrs":{}}
+{"metric_name":"cpu_percent_promql_subquery_no_step","labels":{"host":"server-01","cpu":"cpu0"},"timestamp":"2026-01-29T10:03:00+00:00","value":15,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false,"flags":0,"description":"","unit":"","env":"default","resource_attrs":{},"scope_attrs":{}}
+{"metric_name":"cpu_percent_promql_subquery_no_step","labels":{"host":"server-01","cpu":"cpu0"},"timestamp":"2026-01-29T10:04:00+00:00","value":15,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false,"flags":0,"description":"","unit":"","env":"default","resource_attrs":{},"scope_attrs":{}}
+{"metric_name":"cpu_percent_promql_subquery_no_step","labels":{"host":"server-01","cpu":"cpu0"},"timestamp":"2026-01-29T10:05:00+00:00","value":15,"temporality":"Unspecified","type_":"Gauge","is_monotonic":false,"flags":0,"description":"","unit":"","env":"default","resource_attrs":{},"scope_attrs":{}}

tests/integration/testdata/alerts/test_scenarios/promql_subquery_no_step/rule.json

@@ -0,0 +1,58 @@
+{
+  "alert": "promql_subquery_no_step",
+  "ruleType": "promql_rule",
+  "alertType": "METRIC_BASED_ALERT",
+  "condition": {
+    "thresholds": {
+      "kind": "basic",
+      "spec": [
+        {
+          "name": "critical",
+          "target": 10,
+          "matchType": "at_least_once",
+          "op": "above",
+          "channels": [
+            "test channel"
+          ]
+        }
+      ]
+    },
+    "compositeQuery": {
+      "queryType": "promql",
+      "panelType": "graph",
+      "queries": [
+        {
+          "type": "promql",
+          "spec": {
+            "name": "A",
+            "query": "max_over_time({\"cpu_percent_promql_subquery_no_step\"}[2m:])"
+          }
+        }
+      ]
+    },
+    "selectedQueryName": "A"
+  },
+  "evaluation": {
+    "kind": "rolling",
+    "spec": {
+      "evalWindow": "5m0s",
+      "frequency": "15s"
+    }
+  },
+  "labels": {},
+  "annotations": {
+    "description": "This alert is fired when the defined metric (current value: {{$value}}) crosses the threshold ({{$threshold}})",
+    "summary": "This alert is fired when the defined metric (current value: {{$value}}) crosses the threshold ({{$threshold}})"
+  },
+  "notificationSettings": {
+    "groupBy": [],
+    "usePolicy": false,
+    "renotify": {
+      "enabled": false,
+      "interval": "30m",
+      "alertStates": []
+    }
+  },
+  "version": "v5",
+  "schemaVersion": "v2alpha1"
+}

tests/integration/tests/alerts/04_promql_subquery_no_step.py

@@ -0,0 +1,93 @@
+import json
+import uuid
+from collections.abc import Callable
+from datetime import UTC, datetime, timedelta
+
+from wiremock.client import HttpMethods, Mapping, MappingRequest, MappingResponse
+
+from fixtures import types
+from fixtures.alerts import (
+    update_rule_channel_name,
+    verify_webhook_alert_expectation,
+)
+from fixtures.fs import get_testdata_file_path
+
+TEST_CASE = types.AlertTestCase(
+    name="promql_subquery_no_step",
+    rule_path="alerts/test_scenarios/promql_subquery_no_step/rule.json",
+    alert_data=[
+        types.AlertData(
+            type="metrics",
+            data_path="alerts/test_scenarios/promql_subquery_no_step/alert_data.jsonl",
+        ),
+    ],
+    alert_expectation=types.AlertExpectation(
+        should_alert=True,
+        wait_time_seconds=30,
+        expected_alerts=[
+            types.FiringAlert(
+                labels={
+                    "alertname": "promql_subquery_no_step",
+                    "threshold.name": "critical",
+                }
+            ),
+        ],
+    ),
+)
+
+
+def test_promql_rule_subquery_without_step(
+    notification_channel: types.TestContainerDocker,
+    make_http_mocks: Callable[[types.TestContainerDocker, list[Mapping]], None],
+    create_webhook_notification_channel: Callable[[str, str, dict, bool], str],
+    create_alert_rule: Callable[[dict], str],
+    insert_alert_data: Callable[[list[types.AlertData], datetime], None],
+):
+    """
+    A promql rule with a step-less subquery ([2m:]) must evaluate and fire.
+    A nil NoStepSubqueryIntervalFn segfaults the process on first evaluation.
+    """
+    notification_channel_name = str(uuid.uuid4())
+    webhook_endpoint_path = f"/alert/{notification_channel_name}"
+    notification_url = notification_channel.container_configs["8080"].get(webhook_endpoint_path)
+
+    make_http_mocks(
+        notification_channel,
+        [
+            Mapping(
+                request=MappingRequest(
+                    method=HttpMethods.POST,
+                    url=webhook_endpoint_path,
+                ),
+                response=MappingResponse(
+                    status=200,
+                    json_body={},
+                ),
+                persistent=False,
+            )
+        ],
+    )
+
+    create_webhook_notification_channel(
+        channel_name=notification_channel_name,
+        webhook_url=notification_url,
+        http_config={},
+        send_resolved=False,
+    )
+
+    insert_alert_data(
+        TEST_CASE.alert_data,
+        base_time=datetime.now(tz=UTC) - timedelta(minutes=5),
+    )
+
+    rule_path = get_testdata_file_path(TEST_CASE.rule_path)
+    with open(rule_path, encoding="utf-8") as f:
+        rule_data = json.loads(f.read())
+    update_rule_channel_name(rule_data, notification_channel_name)
+    create_alert_rule(rule_data)
+
+    verify_webhook_alert_expectation(
+        notification_channel,
+        notification_channel_name,
+        TEST_CASE.alert_expectation,
+    )

tests/integration/tests/promqlconformance/04_no_step_subquery.py

@@ -0,0 +1,65 @@
+from collections.abc import Callable
+from datetime import UTC, datetime, timedelta
+from http import HTTPStatus
+from uuid import uuid4
+
+from fixtures import types
+from fixtures.auth import USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD
+from fixtures.metrics import Metrics
+from fixtures.querier import get_all_series, make_query_request
+
+MINUTE_MS = 60_000
+
+LEGS: list[tuple[str, dict | None]] = [
+    ("default", None),
+    ("clickhousev2", {"X-SigNoz-PromQL-Provider": "clickhousev2"}),
+]
+
+
+def test_promql_subquery_without_step_evaluates(
+    signoz: types.SigNoz,
+    create_user_admin: None,  # pylint: disable=unused-argument
+    get_token: Callable[[str, str], str],
+    insert_metrics: Callable[[list[Metrics]], None],
+) -> None:
+    """
+    A subquery that omits its step, e.g. `metric[5m:]`, is valid PromQL: the
+    engine fills in its default resolution. A nil NoStepSubqueryIntervalFn
+    segfaults the whole process on the first such query.
+    """
+    end_ms = (int((datetime.now(tz=UTC) - timedelta(minutes=5)).timestamp() * 1000) // MINUTE_MS) * MINUTE_MS
+    start_ms = end_ms - 30 * MINUTE_MS
+
+    metric = f"no_step_subquery_gauge_{uuid4().hex[:8]}"
+    insert_metrics(
+        [
+            Metrics(
+                metric_name=metric,
+                labels={"host": "server-01"},
+                timestamp=datetime.fromtimestamp(ts_ms / 1000, tz=UTC),
+                value=42.0,
+            )
+            for ts_ms in range(start_ms, end_ms + 1, MINUTE_MS)
+        ]
+    )
+
+    token = get_token(USER_ADMIN_EMAIL, USER_ADMIN_PASSWORD)
+
+    for leg, headers in LEGS:
+        query = {"type": "promql", "spec": {"name": "A", "query": f"max_over_time({metric}[5m:])"}}
+        response = make_query_request(signoz, token, start_ms, end_ms, [query], headers=headers)
+        assert response.status_code == HTTPStatus.OK, f"{leg}: {response.text[:300]}"
+        series = get_all_series(response.json(), "A")
+        assert series, f"{leg}: the subquery must return the inserted series"
+        values = {point["value"] for entry in series for point in entry.get("values") or []}
+        assert values == {42.0}, f"{leg}: {sorted(values)[:5]}"
+
+    # A plain follow-up query proves the process survived the subquery legs.
+    response = make_query_request(
+        signoz,
+        token,
+        start_ms,
+        end_ms,
+        [{"type": "promql", "spec": {"name": "A", "query": metric}}],
+    )
+    assert response.status_code == HTTPStatus.OK, response.text[:300]