Add support for rule group labels (#6665)

* Add support for rule group labels

Signed-off-by: Mustafain Ali Khan <mustalik@amazon.com>

* fix test

Signed-off-by: Mustafain Ali Khan <mustalik@amazon.com>

* Include group labels in backup and update integration test

Signed-off-by: Mustafain Ali Khan <mustalik@amazon.com>

---------

Signed-off-by: Mustafain Ali Khan <mustalik@amazon.com>

Author: mustafain117

CHANGELOG.md

@@ -8,6 +8,7 @@
 * [FEATURE] Ingester/StoreGateway: Add `ResourceMonitor` module in Cortex, and add `ResourceBasedLimiter` in Ingesters and StoreGateways. #6674
 * [FEATURE] Ingester: Support out-of-order native histogram ingestion. It automatically enabled when `-ingester.out-of-order-time-window > 0` and `-blocks-storage.tsdb.enable-native-histograms=true`. #6626 #6663
 * [FEATURE] Ruler: Add support for percentage based sharding for rulers. #6680
+* [FEATURE] Ruler: Add support for group labels. #6665
 * [ENHANCEMENT] Querier: Support query parameters to metadata api (/api/v1/metadata) to allow user to limit metadata to return. #6681
 * [ENHANCEMENT] Ingester: Add a `cortex_ingester_active_native_histogram_series` metric to track # of active NH series. #6695
 * [ENHANCEMENT] Query Frontend: Add new limit `-frontend.max-query-response-size` for total query response size after decompression in query frontend. #6607

integration/ruler_test.go

@@ -311,6 +311,16 @@ func testRulerAPIWithSharding(t *testing.T, enableRulesBackup bool) {
 	expectedNames := make([]string, numRulesGroups)
 	alertCount := 0
 	evalInterval, _ := model.ParseDuration("1s")
+	groupLabels := map[string]string{
+		"group_label_1":   "val1",
+		"group_label_2":   "val2",
+		"duplicate_label": "group_val",
+	}
+	ruleLabels := map[string]string{
+		"rule_label_1":    "val1",
+		"rule_label_2":    "val2",
+		"duplicate_label": "rule_val",
+	}
 	for i := 0; i < numRulesGroups; i++ {
 		num := random.Intn(100)
 		var ruleNode yaml.Node
@@ -319,26 +329,28 @@ func testRulerAPIWithSharding(t *testing.T, enableRulesBackup bool) {
 		ruleNode.SetString(fmt.Sprintf("rule_%d", i))
 		exprNode.SetString(strconv.Itoa(i))
 		ruleName := fmt.Sprintf("test_%d", i)
-
 		expectedNames[i] = ruleName
 		if num%2 == 0 {
 			alertCount++
 			ruleGroups[i] = rulefmt.RuleGroup{
 				Name:     ruleName,
 				Interval: evalInterval,
 				Rules: []rulefmt.RuleNode{{
-					Alert: ruleNode,
-					Expr:  exprNode,
+					Alert:  ruleNode,
+					Expr:   exprNode,
+					Labels: ruleLabels,
 				}},
+				Labels: groupLabels,
 			}
 		} else {
 			ruleGroups[i] = rulefmt.RuleGroup{
 				Name:     ruleName,
 				Interval: evalInterval,
 				Rules: []rulefmt.RuleNode{{
 					Record: ruleNode,
-					Expr:   exprNode,
+					Labels: ruleLabels,
 				}},
+				Labels: groupLabels,
 			}
 		}
 	}
@@ -487,6 +499,32 @@ func testRulerAPIWithSharding(t *testing.T, enableRulesBackup bool) {
 				assert.Greater(t, alertsCount, 0, "Expected greater than 0 alerts but got %d", alertsCount)
 			},
 		},
+		"Filter Rules and verify Group Labels exist": {
+			filter: e2ecortex.RuleFilter{
+				RuleType: "alert",
+			},
+			resultCheckFn: func(t assert.TestingT, ruleGroups []*ruler.RuleGroup) {
+				for _, ruleGroup := range ruleGroups {
+					rule := ruleGroup.Rules[0].(map[string]interface{})
+					ruleType := rule["type"]
+					assert.Equal(t, "alerting", ruleType, "Expected 'alerting' rule type but got %s", ruleType)
+					responseJson, err := json.Marshal(rule)
+					assert.NoError(t, err)
+					ar := &alertingRule{}
+					assert.NoError(t, json.Unmarshal(responseJson, ar))
+					if !ar.LastEvaluation.IsZero() {
+						// Labels will be merged only if groups are loaded to Prometheus rule manager
+						assert.Equal(t, 5, len(ar.Labels))
+					}
+					for _, label := range ar.Labels {
+						if label.Name == "duplicate_label" {
+							// rule label should override group label
+							assert.Equal(t, ruleLabels["duplicate_label"], label.Value)
+						}
+					}
+				}
+			},
+		},
 	}
 	// For each test case, fetch the rules with configured filters, and ensure the results match.
 	if enableRulesBackup {

pkg/ruler/ruler.go

@@ -1257,6 +1257,7 @@ func (r *Ruler) ruleGroupListToGroupStateDesc(userID string, backupGroups rulesp
 				User:        userID,
 				Limit:       group.Limit,
 				QueryOffset: group.QueryOffset,
+				Labels:      group.Labels,
 			},
 			// We are keeping default value for EvaluationTimestamp and EvaluationDuration since the backup is not evaluating
 		}

pkg/ruler/rulespb/compat.go

@@ -26,6 +26,7 @@ func ToProto(user string, namespace string, rl rulefmt.RuleGroup) *RuleGroupDesc
 		User:        user,
 		Limit:       int64(rl.Limit),
 		QueryOffset: queryOffset,
+		Labels:      cortexpb.FromLabelsToLabelAdapters(labels.FromMap(rl.Labels)),
 	}
 	return &rg
 }
@@ -60,6 +61,7 @@ func FromProto(rg *RuleGroupDesc) rulefmt.RuleGroup {
 		Rules:       make([]rulefmt.RuleNode, len(rg.GetRules())),
 		Limit:       int(rg.Limit),
 		QueryOffset: queryOffset,
+		Labels:      cortexpb.FromLabelAdaptersToLabels(rg.Labels).Map(),
 	}
 
 	for i, rl := range rg.GetRules() {

pkg/ruler/rulespb/compat_test.go

@@ -35,6 +35,7 @@ func TestProto(t *testing.T) {
 		Rules:       rules,
 		Interval:    model.Duration(time.Minute),
 		QueryOffset: &queryOffset,
+		Labels:      map[string]string{},
 	}
 
 	desc := ToProto("test", "namespace", rg)

pkg/ruler/rulespb/rules.pb.go

@@ -30,6 +30,10 @@ message RuleGroupDesc {
   int64 limit =10;
   google.protobuf.Duration queryOffset = 11
       [(gogoproto.nullable) = true, (gogoproto.stdduration) = true];
+  repeated cortexpb.LabelPair labels = 12 [
+    (gogoproto.nullable) = false,
+    (gogoproto.customtype) = "github.com/cortexproject/cortex/pkg/cortexpb.LabelAdapter"
+  ];
 }
 
 // RuleDesc is a proto representation of a Prometheus Rule