Add cluster ILM metrics in metrics-v3

Signed-off-by: Bala.FA <[email protected]>

cmd/metrics-v2.go

@@ -273,13 +273,10 @@ const (
  vmemory = "virtual_memory_bytes"
  cpu = "cpu_total_seconds"
 
- expiryPendingTasks MetricName = "expiry_pending_tasks"
  expiryMissedTasks MetricName = "expiry_missed_tasks"
  expiryMissedFreeVersions MetricName = "expiry_missed_freeversions"
  expiryMissedTierJournalTasks MetricName = "expiry_missed_tierjournal_tasks"
  expiryNumWorkers MetricName = "expiry_num_workers"
- transitionPendingTasks MetricName = "transition_pending_tasks"
- transitionActiveTasks MetricName = "transition_active_tasks"
  transitionMissedTasks MetricName = "transition_missed_immediate_tasks"
 
  transitionedBytes MetricName = "transitioned_bytes"

cmd/metrics-v3-cluster-ilm.go

@@ -0,0 +1,55 @@
+// Copyright (c) 2024 MinIO, Inc.
+//
+// # This file is part of MinIO Object Storage stack
+//
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Affero General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// This program is distributed in the hope that it will be useful
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+// GNU Affero General Public License for more details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+package cmd
+
+import (
+ "context"
+)
+
+const (
+ expiryPendingTasks = "expiry_pending_tasks"
+ transitionActiveTasks = "transition_active_tasks"
+ transitionMissedImmediateTasks = "transition_missed_immediate_tasks"
+ transitionPendingTasks = "transition_pending_tasks"
+ versionsScanned = "versions_scanned"
+)
+
+var (
+ ilmExpiryPendingTasksMD = NewCounterMD(expiryPendingTasks, "Number of pending ILM expiry tasks in the queue")
+ ilmTransitionActiveTasksMD = NewCounterMD(transitionActiveTasks, "Number of active ILM transition tasks")
+ ilmTransitionMissedImmediateTasksMD = NewCounterMD(transitionMissedImmediateTasks, "Number of missed immediate ILM transition tasks")
+ ilmTransitionPendingTasksMD = NewCounterMD(transitionPendingTasks, "Number of pending ILM transition tasks in the queue")
+ ilmVersionsScannedMD = NewCounterMD(versionsScanned, "")
+)
+
+// loadClusterILMMetrics - `MetricsLoaderFn` for cluster ILM metrics.
+func loadClusterILMMetrics(_ context.Context, m MetricValues, _ *metricsCache) error {
+ if globalExpiryState != nil {
+ m.Set(expiryPendingTasks, float64(globalExpiryState.PendingTasks()))
+ }
+ if globalTransitionState != nil {
+ m.Set(transitionActiveTasks, float64(globalTransitionState.ActiveTasks()))
+ m.Set(transitionMissedImmediateTasks, float64(globalTransitionState.MissedImmediateTasks()))
+ m.Set(transitionPendingTasks, float64(globalTransitionState.PendingTasks()))
+ }
+
+ // FIXME: add version scanned
+ // m.Set(versionsScanned, 0)
+
+ return nil
+}

cmd/metrics-v3.go

@@ -44,6 +44,7 @@ const (
  clusterUsageBucketsCollectorPath collectorPath = "/cluster/usage/buckets"
  clusterErasureSetCollectorPath collectorPath = "/cluster/erasure-set"
  clusterAuditCollectorPath collectorPath = "/cluster/audit"
+ clusterILMCollectorPath collectorPath = "/cluster/ilm"
 )
 
 const (
@@ -228,6 +229,17 @@ func newMetricGroups(r *prometheus.Registry) *metricsV3Collection {
  loadClusterAuditMetrics,
  )
 
+ clusterILMMG := NewMetricsGroup(clusterILMCollectorPath,
+ []MetricDescriptor{
+ ilmExpiryPendingTasksMD,
+ ilmTransitionActiveTasksMD,
+ ilmTransitionMissedImmediateTasksMD,
+ ilmTransitionPendingTasksMD,
+ ilmVersionsScannedMD,
+ },
+ loadClusterILMMetrics,
+ )
+
  allMetricGroups := []*MetricsGroup{
  apiRequestsMG,
  apiBucketMG,
@@ -241,6 +253,7 @@ func newMetricGroups(r *prometheus.Registry) *metricsV3Collection {
  clusterUsageBucketsMG,
  clusterErasureSetMG,
  clusterAuditMG,
+ clusterILMMG,
  }
 
  // Bucket metrics are special, they always include the bucket label. These

docs/metrics/v3.md

@@ -209,3 +209,13 @@ The standard metrics groups for ProcessCollector and GoCollector are not shown b
 | `minio_cluster_erasure_set_online_drives_count` | `gauge` | Count of online drives in the erasure set in a pool | `pool_id,set_id` |
 | `minio_cluster_erasure_set_healing_drives_count` | `gauge` | Count of healing drives in the erasure set in a pool | `pool_id,set_id` |
 | `minio_cluster_erasure_set_health` | `gauge` | Health of the erasure set in a pool (1=healthy, 0=unhealthy) | `pool_id,set_id` |
+
+### `/cluster/ilm`
+
+| Name | Type | Help | Labels |
+|-------------------------------------------------------|-----------|-----------------------------------------------------|--------|
+| `minio_cluster_ilm_expiry_pending_tasks` | `counter` | Number of pending ILM expiry tasks in the queue | |
+| `minio_cluster_ilm_transition_active_tasks` | `counter` | Number of active ILM transition tasks | |
+| `minio_cluster_ilm_transition_missed_immediate_tasks` | `counter` | Number of missed immediate ILM transition tasks | |
+| `minio_cluster_ilm_transition_pending_tasks` | `counter` | Number of pending ILM transition tasks in the queue | |
+| `minio_cluster_ilm_versions_scanned` | `counter` | | |