diff --git a/changelog/23.0/23.0.0/summary.md b/changelog/23.0/23.0.0/summary.md index d8cd8f8a2c7..0b8c1b9bb2e 100644 --- a/changelog/23.0/23.0.0/summary.md +++ b/changelog/23.0/23.0.0/summary.md @@ -3,6 +3,8 @@ ### Table of Contents - **[Minor Changes](#minor-changes)** + - **[Deprecations](#deprecations)** + - [Metrics](#deprecated-metrics) - **[Deletions](#deletions)** - [Metrics](#deleted-metrics) - **[New Metrics](#new-metrics)** @@ -17,6 +19,14 @@ ## Minor Changes +### Deprecations + +#### Metrics + +| Component | Metric Name | Notes | Deprecation PR | +|:---------:|:-------------------------:|:--------------------------------------:|:-------------------------------------------------------:| +| `vtorc` | `DiscoverInstanceTimings` | Replaced by `DiscoveryInstanceTimings` | [#18406](https://github.com/vitessio/vitess/pull/18406) | + ### Deletions #### Metrics diff --git a/go/test/endtoend/vtorc/api/api_test.go b/go/test/endtoend/vtorc/api/api_test.go index 22091e5cce5..b4f1ca70b40 100644 --- a/go/test/endtoend/vtorc/api/api_test.go +++ b/go/test/endtoend/vtorc/api/api_test.go @@ -110,24 +110,49 @@ func TestAPIEndpoints(t *testing.T) { t.Run("Check Vars and Metrics", func(t *testing.T) { utils.CheckVarExists(t, vtorc, "AnalysisChangeWrite") utils.CheckVarExists(t, vtorc, "AuditWrite") + utils.CheckVarExists(t, vtorc, "CurrentErrantGTIDCount") + utils.CheckVarExists(t, vtorc, "DetectedProblems") utils.CheckVarExists(t, vtorc, "DiscoveriesAttempt") utils.CheckVarExists(t, vtorc, "DiscoveriesFail") utils.CheckVarExists(t, vtorc, "DiscoveriesInstancePollSecondsExceeded") utils.CheckVarExists(t, vtorc, "DiscoveriesQueueLength") utils.CheckVarExists(t, vtorc, "DiscoveriesRecentCount") + utils.CheckVarExists(t, vtorc, "DiscoveryWorkers") + utils.CheckVarExists(t, vtorc, "DiscoveryWorkersActive") + utils.CheckVarExists(t, vtorc, "DiscoveryInstanceTimings") + utils.CheckVarExists(t, vtorc, "FailedRecoveries") utils.CheckVarExists(t, vtorc, "InstanceRead") utils.CheckVarExists(t, vtorc, "InstanceReadTopology") + utils.CheckVarExists(t, vtorc, "PendingRecoveries") + utils.CheckVarExists(t, vtorc, "RecoveriesCount") + utils.CheckVarExists(t, vtorc, "ShardLocksActive") + utils.CheckVarExists(t, vtorc, "ShardLockTimings") + utils.CheckVarExists(t, vtorc, "SuccessfulRecoveries") + utils.CheckVarExists(t, vtorc, "TabletsWatchedByCell") + utils.CheckVarExists(t, vtorc, "TabletsWatchedByShard") // Metrics registered in prometheus utils.CheckMetricExists(t, vtorc, "vtorc_analysis_change_write") utils.CheckMetricExists(t, vtorc, "vtorc_audit_write") + utils.CheckMetricExists(t, vtorc, "vtorc_detected_problems") utils.CheckMetricExists(t, vtorc, "vtorc_discoveries_attempt") utils.CheckMetricExists(t, vtorc, "vtorc_discoveries_fail") utils.CheckMetricExists(t, vtorc, "vtorc_discoveries_instance_poll_seconds_exceeded") utils.CheckMetricExists(t, vtorc, "vtorc_discoveries_queue_length") utils.CheckMetricExists(t, vtorc, "vtorc_discoveries_recent_count") + utils.CheckMetricExists(t, vtorc, "vtorc_discovery_instance_timings_bucket") + utils.CheckMetricExists(t, vtorc, "vtorc_discovery_workers") + utils.CheckMetricExists(t, vtorc, "vtorc_discovery_workers_active") + utils.CheckMetricExists(t, vtorc, "vtorc_errant_gtid_tablet_count") utils.CheckMetricExists(t, vtorc, "vtorc_instance_read") utils.CheckMetricExists(t, vtorc, "vtorc_instance_read_topology") + utils.CheckMetricExists(t, vtorc, "vtorc_pending_recoveries") + utils.CheckMetricExists(t, vtorc, "vtorc_recoveries_count") + utils.CheckMetricExists(t, vtorc, "vtorc_shard_locks_active") + utils.CheckMetricExists(t, vtorc, "vtorc_shard_lock_timings_bucket") + utils.CheckMetricExists(t, vtorc, "vtorc_successful_recoveries") + utils.CheckMetricExists(t, vtorc, "vtorc_tablets_watched_by_cell") + utils.CheckMetricExists(t, vtorc, "vtorc_tablets_watched_by_shard") }) t.Run("Disable Recoveries API", func(t *testing.T) { diff --git a/go/vt/vtorc/logic/vtorc.go b/go/vt/vtorc/logic/vtorc.go index 45568cd745c..0d1ab8bd54a 100644 --- a/go/vt/vtorc/logic/vtorc.go +++ b/go/vt/vtorc/logic/vtorc.go @@ -59,8 +59,10 @@ var ( discoveryWorkersGauge = stats.NewGauge("DiscoveryWorkers", "Number of discovery workers") discoveryWorkersActiveGauge = stats.NewGauge("DiscoveryWorkersActive", "Number of discovery workers actively discovering tablets") - discoverInstanceTimingsActions = []string{"Backend", "Instance", "Other"} - discoverInstanceTimings = stats.NewTimings("DiscoverInstanceTimings", "Timings for instance discovery actions", "Action", discoverInstanceTimingsActions...) + discoveryInstanceTimingsActions = []string{"Backend", "Instance", "Other"} + discoveryInstanceTimings = stats.NewTimings("DiscoveryInstanceTimings", "Timings for instance discovery actions", "Action", discoveryInstanceTimingsActions...) + // TODO: deprecate in v24 + discoverInstanceTimings = stats.NewTimings("DiscoverInstanceTimings", "Timings for instance discovery actions", "Action", discoveryInstanceTimingsActions...) ) var discoveryMetrics = collection.CreateOrReturnCollection(DiscoveryMetricsName) @@ -197,10 +199,15 @@ func DiscoverInstance(tabletAlias string, forceDiscovery bool) { instanceLatency := latency.Elapsed("instance") otherLatency := totalLatency - (backendLatency + instanceLatency) + // TODO: deprecate in v24 discoverInstanceTimings.Add("Backend", backendLatency) discoverInstanceTimings.Add("Instance", instanceLatency) discoverInstanceTimings.Add("Other", otherLatency) + discoveryInstanceTimings.Add("Backend", backendLatency) + discoveryInstanceTimings.Add("Instance", instanceLatency) + discoveryInstanceTimings.Add("Other", otherLatency) + if forceDiscovery { log.Infof("Force discovered - %+v, err - %v", instance, err) }