vtorc: add detected_problems counter

go/stats/counters.go

@@ -368,6 +368,11 @@ func NewGaugesWithMultiLabels(name, help string, labels []string) *GaugesWithMul
 	return t
 }
 
+// GetLabelName returns a label name using the provided values.
+func (mg *GaugesWithMultiLabels) GetLabelName(names ...string) string {
+	return safeJoinLabels(names, nil)
+}
+
 // Set sets the value of a named counter.
 // len(names) must be equal to len(Labels).
 func (mg *GaugesWithMultiLabels) Set(names []string, value int64) {
@@ -377,6 +382,17 @@ func (mg *GaugesWithMultiLabels) Set(names []string, value int64) {
 	mg.counters.set(safeJoinLabels(names, nil), value)
 }
 
+// ResetKey resets a specific key.
+//
+// It is the equivalent of `Reset(names)` except that it expects the key to
+// be obtained from the internal counters map.
+//
+// This is useful when you range over all internal counts and you want to reset
+// specific keys.
+func (mg *GaugesWithMultiLabels) ResetKey(key string) {
+	mg.counters.set(key, 0)
+}
+
 // GaugesFuncWithMultiLabels is a wrapper around CountersFuncWithMultiLabels
 // for values that go up/down for implementations (like Prometheus) that
 // need to differ between Counters and Gauges.

go/test/endtoend/vtorc/general/vtorc_test.go

@@ -28,6 +28,7 @@ import (
 	"vitess.io/vitess/go/test/endtoend/cluster"
 	"vitess.io/vitess/go/test/endtoend/vtorc/utils"
 	"vitess.io/vitess/go/vt/log"
+	"vitess.io/vitess/go/vt/vtorc/inst"
 	"vitess.io/vitess/go/vt/vtorc/logic"
 )
 
@@ -102,6 +103,7 @@ func TestKeyspaceShard(t *testing.T) {
 // 3. stop replication, let vtorc repair
 // 4. setup replication from non-primary, let vtorc repair
 // 5. make instance A replicates from B and B from A, wait for repair
+// 6. disable recoveries and make sure the detected problems are set correctly.
 func TestVTOrcRepairs(t *testing.T) {
 	defer utils.PrintVTOrcLogsOnFailure(t, clusterInfo.ClusterInstance)
 	defer cluster.PanicHandler(t)
@@ -224,6 +226,39 @@ func TestVTOrcRepairs(t *testing.T) {
 		utils.WaitForTabletType(t, replica, "drained")
 	})
 
+	t.Run("Sets DetectedProblems metric correctly", func(t *testing.T) {
+		// Since we're using a boolean metric here, disable recoveries for now.
+		status, _, err := utils.MakeAPICall(t, vtOrcProcess, "/api/disable-global-recoveries")
+		require.NoError(t, err)
+		require.Equal(t, 200, status)
+
+		// Make the current primary database read-only.
+		_, err = utils.RunSQL(t, "set global read_only=ON", curPrimary, "")
+		require.NoError(t, err)
+
+		// Wait for problems to be set.
+		utils.WaitForDetectedProblems(t, vtOrcProcess,
+			string(inst.PrimaryIsReadOnly),
+			curPrimary.Alias,
+			keyspace.Name,
+			shard0.Name,
+			1,
+		)
+
+		// Enable recoveries.
+		status, _, err = utils.MakeAPICall(t, vtOrcProcess, "/api/enable-global-recoveries")
+		require.NoError(t, err)
+		assert.Equal(t, 200, status)
+
+		// wait for detected problem to be cleared.
+		utils.WaitForDetectedProblems(t, vtOrcProcess,
+			string(inst.PrimaryIsReadOnly),
+			curPrimary.Alias,
+			keyspace.Name,
+			shard0.Name,
+			0,
+		)
+	})
 }
 
 func TestRepairAfterTER(t *testing.T) {

go/test/endtoend/vtorc/utils/utils.go

@@ -969,6 +969,39 @@ func WaitForSuccessfulRecoveryCount(t *testing.T, vtorcInstance *cluster.VTOrcPr
 	assert.EqualValues(t, countExpected, successCount)
 }
 
+// WaitForDetectedProblems waits until the given analysis code, alias, keyspace and shard count matches the count expected.
+func WaitForDetectedProblems(t *testing.T, vtorcInstance *cluster.VTOrcProcess, code, alias, ks, shard string, expect int) {
+	t.Helper()
+	key := strings.Join([]string{code, alias, ks, shard}, ".")
+	timeout := 15 * time.Second
+	startTime := time.Now()
+
+	for time.Since(startTime) < timeout {
+		vars := vtorcInstance.GetVars()
+		problems := vars["DetectedProblems"].(map[string]interface{})
+		actual := problems[key]
+		if actual == expect {
+			return
+		}
+		time.Sleep(time.Second)
+	}
+
+	vars := vtorcInstance.GetVars()
+	problems := vars["DetectedProblems"].(map[string]interface{})
+	actual, ok := problems[key]
+
+	assert.True(t, ok,
+		"The metric DetectedProblems[%s] should exist but does not (all problems: %+v)",
+		key, problems,
+	)
+
+	assert.EqualValues(t, expect, actual,
+		"The metric DetectedProblems[%s] should be %v but is %v (all problems: %+v)",
+		key, expect, actual,
+		problems,
+	)
+}
+
 // WaitForTabletType waits for the tablet to reach a certain type.
 func WaitForTabletType(t *testing.T, tablet *cluster.Vttablet, expectedTabletType string) {
 	t.Helper()

go/vt/vtorc/inst/analysis.go

@@ -25,7 +25,6 @@ import (
 )
 
 type AnalysisCode string
-type StructureAnalysisCode string
 
 const (
 	NoProblem                              AnalysisCode = "NoProblem"
@@ -61,6 +60,8 @@ const (
 	ErrantGTIDDetected                     AnalysisCode = "ErrantGTIDDetected"
 )
 
+type StructureAnalysisCode string
+
 const (
 	StatementAndMixedLoggingReplicasStructureWarning     StructureAnalysisCode = "StatementAndMixedLoggingReplicasStructureWarning"
 	StatementAndRowLoggingReplicasStructureWarning       StructureAnalysisCode = "StatementAndRowLoggingReplicasStructureWarning"

go/vt/vtorc/logic/topology_recovery.go

@@ -63,6 +63,17 @@ var (
 
 	countPendingRecoveries = stats.NewGauge("PendingRecoveries", "Count of the number of pending recoveries")
 
+	// detectedProblems is used to track the number of detected problems.
+	//
+	// When an issue is active it will be set to 1, when it is no longer active
+	// it will be reset back to 0.
+	detectedProblems = stats.NewGaugesWithMultiLabels("DetectedProblems", "Count of the different detected problems", []string{
+		"Analysis",
+		"TabletAlias",
+		"Keyspace",
+		"Shard",
+	})
+
 	// recoveriesCounter counts the number of recoveries that VTOrc has performed
 	recoveriesCounter = stats.NewCountersWithSingleLabel("RecoveriesCount", "Count of the different recoveries performed", "RecoveryType", actionableRecoveriesNames...)
 
@@ -755,17 +766,42 @@ func CheckAndRecover() {
 		log.Error(err)
 		return
 	}
+
+	// Regardless of if the problem is solved or not we want to monitor active
+	// issues, we use a map of labels and set a counter to `1` for each problem
+	// then we reset any counter that is not present in the current analysis.
+	active := make(map[string]struct{})
+	for _, e := range replicationAnalysis {
+		if e.Analysis != inst.NoProblem {
+			names := [...]string{
+				string(e.Analysis),
+				e.AnalyzedInstanceAlias,
+				e.AnalyzedKeyspace,
+				e.AnalyzedShard,
+			}
+
+			key := detectedProblems.GetLabelName(names[:]...)
+			active[key] = struct{}{}
+			detectedProblems.Set(names[:], 1)
+		}
+	}
+
+	// Reset any non-active problems.
+	for key := range detectedProblems.Counts() {
+		if _, ok := active[key]; !ok {
+			detectedProblems.ResetKey(key)
+		}
+	}
+
 	// intentionally iterating entries in random order
 	for _, j := range rand.Perm(len(replicationAnalysis)) {
 		analysisEntry := replicationAnalysis[j]
 
 		go func() {
-			err = executeCheckAndRecoverFunction(analysisEntry)
-			if err != nil {
+			if err := executeCheckAndRecoverFunction(analysisEntry); err != nil {
 				log.Error(err)
 			}
 		}()
-
 	}
 }