Skip to content

vtorc: add keyspace/shard labels to recoveries stats #18304

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
GuptaManan100 merged 6 commits into from
Jun 27, 2025
Merged

vtorc: add keyspace/shard labels to recoveries stats #18304

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
e971935
`vtorc`: add keyspace/shard labels to recoveries stats
timvaillancourt May 27, 2025
956c261
add keyspace/shard to `utils.WaitForSuccessfulRecoveryCount`
timvaillancourt May 27, 2025
a6c5f32
missing updates
timvaillancourt May 27, 2025
2272eb6
Merge remote-tracking branch 'origin/main' into vtorc-recoveries-labels
timvaillancourt Jun 11, 2025
8c18be9
add to changelog
timvaillancourt Jun 11, 2025
f307bb6
add extra note
timvaillancourt Jun 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 14 additions & 0 deletions changelog/23.0/23.0.0/summary.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,8 @@
- [VTGate](#new-vtgate-metrics)
- **[Topology](#minor-changes-topo)**
- [`--consul_auth_static_file` requires 1 or more credentials](#consul_auth_static_file-check-creds)
- **[VTOrc](#minor-changes-vtorc)**
- [Recovery stats to include keyspace/shard](#recoveries-stats-keyspace-shard)
- **[VTTablet](#minor-changes-vttablet)**
- [CLI Flags](#flags-vttablet)
- [Managed MySQL configuration defaults to caching-sha2-password](#mysql-caching-sha2-password)
Expand Down Expand Up @@ -40,6 +42,18 @@

The `--consul_auth_static_file` flag used in several components now requires that 1 or more credentials can be loaded from the provided json file.

### <a id="minor-changes-vtorc"/>VTOrc</a>

#### <a id="recoveries-stats-keyspace-shard">Recovery stats to include keyspace/shard</a>

The following recovery-related stats now include labels for keyspaces and shards:
1. `FailedRecoveries`
2. `PendingRecoveries`
3. `RecoveriesCount`
4. `SuccessfulRecoveries`

Previous to this release, only the recovery "type" was included in labels.

### <a id="minor-changes-vttablet"/>VTTablet</a>

#### <a id="flags-vttablet"/>CLI Flags</a>
Expand Down
38 changes: 19 additions & 19 deletions go/test/endtoend/vtorc/general/vtorc_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,7 +76,7 @@ func TestErrantGTIDOnPreviousPrimary(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

var replica, otherReplica *cluster.Vttablet
Expand Down Expand Up @@ -135,7 +135,7 @@ func TestSingleKeyspace(t *testing.T) {

utils.CheckPrimaryTablet(t, clusterInfo, shard0.Vttablets[0], true)
utils.CheckReplication(t, clusterInfo, shard0.Vttablets[0], shard0.Vttablets[1:], 10*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, clusterInfo.ClusterInstance.VTOrcProcesses[0], logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, clusterInfo.ClusterInstance.VTOrcProcesses[0], logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, clusterInfo.ClusterInstance.VTOrcProcesses[0], keyspace.Name, shard0.Name, 1)
}

Expand All @@ -153,7 +153,7 @@ func TestKeyspaceShard(t *testing.T) {

utils.CheckPrimaryTablet(t, clusterInfo, shard0.Vttablets[0], true)
utils.CheckReplication(t, clusterInfo, shard0.Vttablets[0], shard0.Vttablets[1:], 10*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, clusterInfo.ClusterInstance.VTOrcProcesses[0], logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, clusterInfo.ClusterInstance.VTOrcProcesses[0], logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, clusterInfo.ClusterInstance.VTOrcProcesses[0], keyspace.Name, shard0.Name, 1)
}

Expand All @@ -176,7 +176,7 @@ func TestVTOrcRepairs(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

var replica, otherReplica *cluster.Vttablet
Expand Down Expand Up @@ -204,7 +204,7 @@ func TestVTOrcRepairs(t *testing.T) {
// wait for repair
match := utils.WaitForReadOnlyValue(t, curPrimary, 0)
require.True(t, match)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
})

t.Run("ReplicaReadWrite", func(t *testing.T) {
Expand All @@ -215,7 +215,7 @@ func TestVTOrcRepairs(t *testing.T) {
// wait for repair
match := utils.WaitForReadOnlyValue(t, replica, 1)
require.True(t, match)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 1)
})

t.Run("StopReplication", func(t *testing.T) {
Expand All @@ -225,23 +225,23 @@ func TestVTOrcRepairs(t *testing.T) {

// check replication is setup correctly
utils.CheckReplication(t, clusterInfo, curPrimary, []*cluster.Vttablet{replica, otherReplica}, 15*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 2)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 2)

// Stop just the IO thread on the replica
_, err = utils.RunSQL(t, "STOP REPLICA IO_THREAD", replica, "")
require.NoError(t, err)

// check replication is setup correctly
utils.CheckReplication(t, clusterInfo, curPrimary, []*cluster.Vttablet{replica, otherReplica}, 15*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 3)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 3)

// Stop just the SQL thread on the replica
_, err = utils.RunSQL(t, "STOP REPLICA SQL_THREAD", replica, "")
require.NoError(t, err)

// check replication is setup correctly
utils.CheckReplication(t, clusterInfo, curPrimary, []*cluster.Vttablet{replica, otherReplica}, 15*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 4)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 4)
})

t.Run("ReplicationFromOtherReplica", func(t *testing.T) {
Expand All @@ -257,7 +257,7 @@ func TestVTOrcRepairs(t *testing.T) {

// wait until the source port is set back correctly by vtorc
utils.CheckSourcePort(t, replica, curPrimary, 15*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 5)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 5)

// check that writes succeed
utils.VerifyWritesSucceed(t, clusterInfo, curPrimary, []*cluster.Vttablet{replica, otherReplica}, 15*time.Second)
Expand All @@ -269,7 +269,7 @@ func TestVTOrcRepairs(t *testing.T) {

// wait until heart beat interval has been fixed by vtorc.
utils.CheckHeartbeatInterval(t, replica, 16.5, 15*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 6)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 6)

// check that writes succeed
utils.VerifyWritesSucceed(t, clusterInfo, curPrimary, []*cluster.Vttablet{replica, otherReplica}, 15*time.Second)
Expand All @@ -292,7 +292,7 @@ func TestVTOrcRepairs(t *testing.T) {
// wait for repair
err = utils.WaitForReplicationToStop(t, curPrimary)
require.NoError(t, err)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverPrimaryHasPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverPrimaryHasPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
// check that the writes still succeed
utils.VerifyWritesSucceed(t, clusterInfo, curPrimary, []*cluster.Vttablet{replica, otherReplica}, 10*time.Second)
})
Expand Down Expand Up @@ -515,7 +515,7 @@ func TestVTOrcWithPrs(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

// find any replica tablet other than the current primary
Expand All @@ -541,13 +541,13 @@ func TestVTOrcWithPrs(t *testing.T) {
// check that the replica gets promoted
utils.CheckPrimaryTablet(t, clusterInfo, replica, true)
// Verify that VTOrc didn't run any other recovery
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, keyspace.Name, shard0.Name, 0)
utils.WaitForSuccessfulERSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixPrimaryRecoveryName, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverPrimaryHasPrimaryRecoveryName, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixPrimaryRecoveryName, keyspace.Name, shard0.Name, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.FixReplicaRecoveryName, keyspace.Name, shard0.Name, 0)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverPrimaryHasPrimaryRecoveryName, keyspace.Name, shard0.Name, 0)
utils.VerifyWritesSucceed(t, clusterInfo, replica, shard0.Vttablets, 10*time.Second)
}

Expand Down Expand Up @@ -674,7 +674,7 @@ func TestFullStatusConnectionPooling(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

// Kill the current primary.
Expand Down
14 changes: 7 additions & 7 deletions go/test/endtoend/vtorc/primaryfailure/primary_failure_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@ func TestDownPrimary(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

// find the replica and rdonly tablets
Expand Down Expand Up @@ -100,7 +100,7 @@ func TestDownPrimary(t *testing.T) {

// also check that the replication is working correctly after failover
utils.VerifyWritesSucceed(t, clusterInfo, replica, []*cluster.Vttablet{crossCellReplica}, 10*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulERSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)
t.Run("Check ERS and PRS Vars and Metrics", func(t *testing.T) {
utils.CheckVarExists(t, vtOrcProcess, "EmergencyReparentCounts")
Expand Down Expand Up @@ -165,7 +165,7 @@ func TestDownPrimaryBeforeVTOrc(t *testing.T) {

// also check that the replication is working correctly after failover
utils.VerifyWritesSucceed(t, clusterInfo, replica, []*cluster.Vttablet{rdonly}, 10*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulERSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)
}

Expand All @@ -179,7 +179,7 @@ func TestDeletedPrimaryTablet(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

// find the replica and rdonly tablets
Expand Down Expand Up @@ -230,7 +230,7 @@ func TestDeletedPrimaryTablet(t *testing.T) {

// also check that the replication is working correctly after failover
utils.VerifyWritesSucceed(t, clusterInfo, replica, []*cluster.Vttablet{rdonly}, 10*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverPrimaryTabletDeletedRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverPrimaryTabletDeletedRecoveryName, keyspace.Name, shard0.Name, 1)
}

// TestDeadPrimaryRecoversImmediately test Vtorc ability to recover immediately if primary is dead.
Expand All @@ -250,7 +250,7 @@ func TestDeadPrimaryRecoversImmediately(t *testing.T) {
curPrimary := utils.ShardPrimaryTablet(t, clusterInfo, keyspace, shard0)
assert.NotNil(t, curPrimary, "should have elected a primary")
vtOrcProcess := clusterInfo.ClusterInstance.VTOrcProcesses[0]
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.ElectNewPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulPRSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

// find the replica and rdonly tablets
Expand Down Expand Up @@ -287,7 +287,7 @@ func TestDeadPrimaryRecoversImmediately(t *testing.T) {
utils.WaitForInstancePollSecondsExceededCount(t, vtOrcProcess, "InstancePollSecondsExceeded", 2, false)
// also check that the replication is working correctly after failover
utils.VerifyWritesSucceed(t, clusterInfo, replica, []*cluster.Vttablet{crossCellReplica}, 10*time.Second)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, 1)
utils.WaitForSuccessfulRecoveryCount(t, vtOrcProcess, logic.RecoverDeadPrimaryRecoveryName, keyspace.Name, shard0.Name, 1)
utils.WaitForSuccessfulERSCount(t, vtOrcProcess, keyspace.Name, shard0.Name, 1)

// Parse log file and find out how much time it took for DeadPrimary to recover.
Expand Down
5 changes: 3 additions & 2 deletions go/test/endtoend/vtorc/utils/utils.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -989,14 +989,15 @@ func WaitForReadOnlyValue(t *testing.T, curPrimary *cluster.Vttablet, expectValu
}

// WaitForSuccessfulRecoveryCount waits until the given recovery name's count of successful runs matches the count expected
func WaitForSuccessfulRecoveryCount(t *testing.T, vtorcInstance *cluster.VTOrcProcess, recoveryName string, countExpected int) {
func WaitForSuccessfulRecoveryCount(t *testing.T, vtorcInstance *cluster.VTOrcProcess, recoveryName, keyspace, shard string, countExpected int) {
t.Helper()
timeout := 15 * time.Second
startTime := time.Now()
mapKey := fmt.Sprintf("%s.%s.%s", recoveryName, keyspace, shard)
for time.Since(startTime) < timeout {
vars := vtorcInstance.GetVars()
successfulRecoveriesMap := vars["SuccessfulRecoveries"].(map[string]interface{})
successCount := GetIntFromValue(successfulRecoveriesMap[recoveryName])
successCount := GetIntFromValue(successfulRecoveriesMap[mapKey])
if successCount == countExpected {
return
}
Expand Down
24 changes: 10 additions & 14 deletions go/vt/vtorc/logic/topology_recovery.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,14 +51,6 @@ const (
)

var (
actionableRecoveriesNames = []string{
Copy link
Contributor Author

@timvaillancourt timvaillancourt May 27, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

stats.NewCountersWithMultiLabels does not support a ...string to init with zero, so this becomes unused

RecoverDeadPrimaryRecoveryName,
RecoverPrimaryHasPrimaryRecoveryName,
ElectNewPrimaryRecoveryName,
FixPrimaryRecoveryName,
FixReplicaRecoveryName,
}

countPendingRecoveries = stats.NewGauge("PendingRecoveries", "Count of the number of pending recoveries")

// detectedProblems is used to track the number of detected problems.
Expand All @@ -75,14 +67,17 @@ var (
// shardsLockCounter is a count of in-flight shard locks. Use atomics to read/update.
shardsLockCounter int64

// recoveriesCounterLabels are labels for grouping the counter based stats for recoveries.
recoveriesCounterLabels = []string{"RecoveryType", "Keyspace", "Shard"}

// recoveriesCounter counts the number of recoveries that VTOrc has performed
recoveriesCounter = stats.NewCountersWithSingleLabel("RecoveriesCount", "Count of the different recoveries performed", "RecoveryType", actionableRecoveriesNames...)
recoveriesCounter = stats.NewCountersWithMultiLabels("RecoveriesCount", "Count of the different recoveries performed", recoveriesCounterLabels)

// recoveriesSuccessfulCounter counts the number of successful recoveries that VTOrc has performed
recoveriesSuccessfulCounter = stats.NewCountersWithSingleLabel("SuccessfulRecoveries", "Count of the different successful recoveries performed", "RecoveryType", actionableRecoveriesNames...)
recoveriesSuccessfulCounter = stats.NewCountersWithMultiLabels("SuccessfulRecoveries", "Count of the different successful recoveries performed", recoveriesCounterLabels)

// recoveriesFailureCounter counts the number of failed recoveries that VTOrc has performed
recoveriesFailureCounter = stats.NewCountersWithSingleLabel("FailedRecoveries", "Count of the different failed recoveries performed", "RecoveryType", actionableRecoveriesNames...)
recoveriesFailureCounter = stats.NewCountersWithMultiLabels("FailedRecoveries", "Count of the different failed recoveries performed", recoveriesCounterLabels)

// shardLockTimings measures the timing of LockShard operations.
shardLockTimingsActions = []string{"Lock", "Unlock"}
Expand Down Expand Up @@ -651,13 +646,14 @@ func executeCheckAndRecoverFunction(analysisEntry *inst.ReplicationAnalysis) (er
return err
}
recoveryName := getRecoverFunctionName(checkAndRecoverFunctionCode)
recoveriesCounter.Add(recoveryName, 1)
recoveryLabels := []string{recoveryName, analysisEntry.AnalyzedKeyspace, analysisEntry.AnalyzedShard}
recoveriesCounter.Add(recoveryLabels, 1)
if err != nil {
logger.Errorf("Failed to recover: %+v", err)
recoveriesFailureCounter.Add(recoveryName, 1)
recoveriesFailureCounter.Add(recoveryLabels, 1)
} else {
logger.Info("Recovery succeeded")
recoveriesSuccessfulCounter.Add(recoveryName, 1)
recoveriesSuccessfulCounter.Add(recoveryLabels, 1)
}
if topologyRecovery == nil {
logger.Error("Topology recovery is nil - recovery might have failed")
Expand Down
Loading