[etcd] Use prometheus metricset for etcd v3 metrics

packages/etcd/_dev/deploy/docker/Dockerfile

@@ -1,2 +1,3 @@
 ARG SERVICE_VERSION=${SERVICE_VERSION:-3.5.1}
 FROM bitnami/etcd:${SERVICE_VERSION}
+HEALTHCHECK --interval=15s --retries=90 CMD curl -f http://localhost:2379/metrics

packages/etcd/changelog.yml

@@ -1,4 +1,9 @@
 # newer versions go on top
+- version: "1.0.0"
+  changes:
+    - description: Use prometheus metricset instead of etcd metricset. Add ECS fields. Add dimensions mappings and metric types.
+      type: enhancement
+      link: https://github.com/elastic/integrations/pull/8438
 - version: 0.6.0
   changes:
     - description: Update the package format_version to 3.0.0.

packages/etcd/data_stream/metrics/_dev/test/system/test-default-config.yml

@@ -1,6 +1,3 @@
 vars:
   hosts:
-    - http://{{Hostname}}:2379
-data_stream:
-  vars:
-    server_status_path: /metrics
+    - http://{{Hostname}}:2379/metrics

packages/etcd/data_stream/metrics/agent/stream/stream.yml.hbs

@@ -1,9 +1,13 @@
-metricsets: ["metrics"]
+metricsets: ["collector"]
 hosts:
 {{#each hosts}}
   - {{this}}
 {{/each}}
+metrics_filters.include:
+{{#each metrics_filters.include}}
+    - {{this}}
+{{/each}}
+metrics_path: {{metrics_path}}
 period: {{period}}
-{{#if server_status_path}}
-server_status_path: {{server_status_path}}
-{{/if}}
+use_types: true
+rate_counters: false

packages/etcd/data_stream/metrics/elasticsearch/ingest_pipeline/default.yml

@@ -0,0 +1,86 @@
+---
+description: Pipeline for processing etcd v3 metrics.
+processors:
+  - set:
+      field: event.module
+      value: etcd
+  - rename:
+      field: prometheus.etcd_mvcc_db_total_size_in_bytes.value
+      target_field: etcd.disk.mvcc_db_total_size.bytes
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_disk_wal_fsync_duration_seconds.histogram
+      target_field: etcd.disk.wal_fsync_duration_seconds.histogram
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_disk_backend_commit_duration_seconds.histogram
+      target_field: etcd.disk.backend_commit_duration_seconds.histogram
+      ignore_missing: true
+  - rename:
+      field: prometheus.go_memstats_alloc_bytes.value
+      target_field: etcd.memory.go_memstats_alloc.bytes
+      ignore_missing: true
+  - rename:
+      field: prometheus.go_memstats_alloc_bytes_total.counter
+      target_field: etcd.memory.go_memstats_alloc.total.bytes
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_network_client_grpc_sent_bytes_total.counter
+      target_field: etcd.network.client_grpc_sent.bytes
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_network_client_grpc_received_bytes_total.counter
+      target_field: etcd.network.client_grpc_received.bytes
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_server_has_leader.value
+      target_field: etcd.server.has_leader.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_server_leader_changes_seen_total.counter
+      target_field: etcd.server.leader_changes.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_server_proposals_committed_total.value
+      target_field: etcd.server.proposals_committed.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_server_proposals_pending.value
+      target_field: etcd.server.proposals_pending.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.etcd_server_proposals_failed_total.counter
+      target_field: etcd.server.proposals_failed.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.grpc_server_started_total.counter
+      target_field: etcd.server.grpc_started.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.grpc_server_handled_total.counter
+      target_field: etcd.server.grpc_handled.count
+      ignore_missing: true
+  - rename:
+      field: prometheus.labels
+      target_field: etcd.labels
+      ignore_missing: true
+      ignore_failure: true
+  - fingerprint:
+      fields: ["etcd.labels"]
+      target_field: "etcd.labels.fingerprint"
+      ignore_missing: true
+  - remove:
+      field: prometheus
+      ignore_missing: true
+  - set:
+      field: event.kind
+      value: pipeline_error
+      if: ctx.error?.message != null
+on_failure:
+  - set:
+      field: error.message
+      value: "{{{_ingest.on_failure_message}}}"
+  - append:
+      field: event.kind
+      value: pipeline_error
+      allow_duplicates: false

packages/etcd/data_stream/metrics/fields/ecs.yml

@@ -1,8 +1,35 @@
-- external: ecs
-  name: host.ip
-- external: ecs
-  name: ecs.version
-- external: ecs
-  name: service.address
-- external: ecs
-  name: service.type
+- name: ecs.version
+  external: ecs
+- name: event.duration
+  external: ecs
+- name: event.kind
+  external: ecs
+- name: service.type
+  external: ecs
+- name: service.address
+  dimension: true
+  external: ecs
+- name: agent.id
+  dimension: true
+  external: ecs
+- name: cloud.account.id
+  dimension: true
+  external: ecs
+- name: cloud.region
+  dimension: true
+  external: ecs
+- name: cloud.availability_zone
+  dimension: true
+  external: ecs
+- name: cloud.instance.id
+  dimension: true
+  external: ecs
+- name: cloud.provider
+  dimension: true
+  external: ecs
+- name: container.id
+  dimension: true
+  external: ecs
+- name: host.name
+  dimension: true
+  external: ecs

packages/etcd/data_stream/metrics/fields/fields.yml

@@ -6,105 +6,78 @@
       fields:
         - name: mvcc_db_total_size.bytes
           type: long
+          metric_type: gauge
           description: |
             Size of stored data at MVCC
-        - name: wal_fsync_duration.ns.bucket.*
-          type: long
+        - name: wal_fsync_duration_seconds.histogram
+          type: histogram
           description: |
             Latency for writing ahead logs to disk
-        - name: wal_fsync_duration.ns.count
-          type: long
-          description: |
-            Write ahead logs count
-        - name: wal_fsync_duration.ns.sum
-          type: long
-          description: |
-            Write ahead logs latency sum
-        - name: backend_commit_duration.ns.bucket.*
-          type: long
+        - name: backend_commit_duration_seconds.histogram
+          type: histogram
           description: |
             Latency for writing backend changes to disk
-        - name: backend_commit_duration.ns.count
+    - name: memory
+      type: group
+      fields:
+        - name: go_memstats_alloc.bytes
           type: long
+          metric_type: gauge
           description: |
-            Backend commits count
-        - name: backend_commit_duration.ns.sum
+            Memory allocated bytes as of MemStats Go
+        - name: go_memstats_alloc.total.bytes
           type: long
+          metric_type: counter
           description: |
-            Backend commits latency sum
-    - name: memory
-      type: group
-      fields:
-        - name: go_memstats_alloc
-          type: group
-          fields:
-            - name: bytes
-              type: long
-              description: |
-                Memory allocated bytes as of MemStats Go
+            Memory allocated bytes as of MemStats Go
     - name: network
       type: group
       fields:
-        - name: client_grpc_sent
-          type: group
-          fields:
-            - name: bytes
-              type: long
-              description: |
-                gRPC sent bytes total
-        - name: client_grpc_received
-          type: group
-          fields:
-            - name: bytes
-              type: long
-              description: |
-                gRPC received bytes total
+        - name: client_grpc_sent.bytes
+          type: long
+          metric_type: counter
+          description: |
+            gRPC sent bytes total
+        - name: client_grpc_received.bytes
+          type: long
+          metric_type: counter
+          description: |
+            gRPC received bytes total
     - name: server
       type: group
       fields:
-        - name: has_leader
-          type: byte
+        - name: has_leader.count
+          type: long
+          metric_type: gauge
           description: |
             Whether a leader exists in the cluster
-        - name: leader_changes
-          type: group
-          fields:
-            - name: count
-              type: long
-              description: |
-                Number of leader changes seen at the cluster
-        - name: proposals_committed
-          type: group
-          fields:
-            - name: count
-              type: long
-              description: |
-                Number of consensus proposals commited
-        - name: proposals_pending
-          type: group
-          fields:
-            - name: count
-              type: long
-              description: |
-                Number of consensus proposals pending
-        - name: proposals_failed
-          type: group
-          fields:
-            - name: count
-              type: long
-              description: |
-                Number of consensus proposals failed
-        - name: grpc_started
-          type: group
-          fields:
-            - name: count
-              type: long
-              description: |
-                Number of sent gRPC requests
-        - name: grpc_handled
-          type: group
-          fields:
-            - name: count
-              type: long
-              description: |
-                Number of received gRPC requests
+        - name: leader_changes.count
+          type: long
+          metric_type: counter
+          description: |
+            Number of leader changes seen at the cluster
+        - name: proposals_committed.count
+          type: long
+          metric_type: gauge
+          description: |
+            Number of consensus proposals commited
+        - name: proposals_pending.count
+          type: long
+          metric_type: gauge
+          description: |
+            Number of consensus proposals pending
+        - name: proposals_failed.count
+          type: long
+          metric_type: counter
+          description: |
+            Number of consensus proposals failed
+        - name: grpc_started.count
+          type: long
+          metric_type: counter
+          description: |
+            Number of sent gRPC requests
+        - name: grpc_handled.count
+          type: long
+          metric_type: counter
+          description: |
+            Number of received gRPC requests

packages/etcd/data_stream/metrics/fields/package-fields.yml

@@ -1,7 +1,13 @@
 - name: etcd
   type: group
   fields:
-    - name: api_version
+    - name: labels.*
       type: keyword
-      description: |
-        Etcd API version for metrics retrieval
+      object_type_mapping_type: "*"
+      description: >-
+        etcd labels.
+    - name: labels.fingerprint
+      type: keyword
+      dimension: true
+      description: >-
+        Unique fingerprint of the etcd labels.

packages/etcd/data_stream/metrics/manifest.yml

@@ -1,7 +1,9 @@
-title: etcd metrics v3
+title: etcd v3 metrics
 type: metrics
 streams:
-  - input: etcd/metrics
+  - input: prometheus/metrics
+    title: etcd v3 metrics
+    description: Collect etcd v3 metrics.
     vars:
       - name: period
         type: text
@@ -10,12 +12,17 @@ streams:
         required: true
         show_user: true
         default: 10s
-      - name: server_status_path
+      - name: metrics_path
         type: text
-        title: Server Status Path
+        title: Metrics Path
         multi: false
-        required: true
-        show_user: false
+        required: false
+        show_user: true
         default: /metrics
-    title: etcd metrics v3
-    description: Collect etcd metrics v3
+      - name: metrics_filters.include
+        type: text
+        title: Metrics Filters Include
+        multi: true
+        required: false
+        show_user: true
+        default: ["etcd_server_has_leader", "etcd_server_leader_changes_seen_total", "etcd_server_proposals_committed_total", "etcd_server_proposals_pending", "etcd_server_proposals_failed_total", "grpc_server_started_total", "grpc_server_handled_total", "etcd_mvcc_db_total_size_in_bytes", "etcd_disk_wal_fsync_duration_seconds", "etcd_disk_backend_commit_duration_seconds", "go_memstats_alloc_bytes", "etcd_network_client_grpc_sent_bytes_total", "etcd_network_client_grpc_received_bytes_total"]