fix(matrix-synapse): add monitoring resources

matrix-synapse/Chart.yaml

@@ -4,9 +4,9 @@ name: matrix-synapse
 description: Matrix reference homeserver
 icon: https://matrix.org/images/matrix-logo.svg
 type: application
-version: 0.0.1
+version: 0.0.2
 # renovate: image=ghcr.io/element-hq/synapse
-appVersion: 1.100.0
+appVersion: 1.101.0
 maintainers:
   - name: WrenIX
     url: https://wrenix.eu

matrix-synapse/README.adoc

@@ -2,9 +2,9 @@
 
 = matrix-synapse
 
-image::https://img.shields.io/badge/Version-0.0.1-informational?style=flat-square[Version: 0.0.1]
+image::https://img.shields.io/badge/Version-0.0.2-informational?style=flat-square[Version: 0.0.2]
 image::https://img.shields.io/badge/Version-application-informational?style=flat-square[Type: application]
-image::https://img.shields.io/badge/AppVersion-1.100.0-informational?style=flat-square[AppVersion: 1.100.0]
+image::https://img.shields.io/badge/AppVersion-1.101.0-informational?style=flat-square[AppVersion: 1.101.0]
 == Maintainers
 
 .Maintainers
@@ -137,6 +137,21 @@ helm uninstall matrix-synapse-release
 | `nil`
 | if set it will overwrite all registry entries
 
+| grafana.dashboards.annotations
+| object
+| `{}`
+|
+
+| grafana.dashboards.enabled
+| bool
+| `false`
+|
+
+| grafana.dashboards.labels.grafana_dashboard
+| string
+| `"1"`
+|
+
 | image.pullPolicy
 | string
 | `"IfNotPresent"`
@@ -272,6 +287,36 @@ helm uninstall matrix-synapse-release
 | `"16Gi"`
 |
 
+| prometheus.podmonitor.enabled
+| bool
+| `false`
+|
+
+| prometheus.podmonitor.labels
+| object
+| `{}`
+|
+
+| prometheus.rules.additionalRules
+| list
+| `[]`
+|
+
+| prometheus.rules.defaults.enabled
+| bool
+| `true`
+|
+
+| prometheus.rules.enabled
+| bool
+| `false`
+|
+
+| prometheus.rules.labels
+| object
+| `{}`
+|
+
 | publicServerName
 | string
 | `nil`

matrix-synapse/grafana_dashboards/README.adoc

@@ -0,0 +1,4 @@
+# Readme
+
+## Changelog from monitoring resources
+- Dashboard: https://github.com/element-hq/synapse/blob/v1.101.0/contrib/grafana/synapse.json

matrix-synapse/templates/configmap_grafana_dashboards.yaml

@@ -0,0 +1,18 @@
+{{- if .Values.grafana.dashboards.enabled }}
+---
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ include "matrix-synapse.fullname" . }}-grafana-db
+  labels:
+    {{- include "matrix-synapse.labels" $ | nindent 4 }}
+    {{- with .Values.grafana.dashboards.labels }}
+    {{- toYaml . | nindent 4 }}
+    {{- end }}
+  {{- with .Values.grafana.dashboards.annotations }}
+  annotations:
+    {{- toYaml . | nindent 4 }}
+  {{- end }}
+data:
+  {{- ($.Files.Glob "grafana_dashboards/*.json" ).AsConfig | nindent 2 }}
+{{- end }}

matrix-synapse/templates/podmonitor.yaml

@@ -0,0 +1,19 @@
+{{- if .Values.prometheus.podmonitor.enabled }}
+apiVersion: monitoring.coreos.com/v1
+kind: PodMonitor
+metadata:
+  name: {{ include "matrix-synapse.fullname" . }}
+  labels:
+    {{- include "matrix-synapse.labels" . | nindent 4 }}
+    {{- with .Values.prometheus.podmonitor.labels }}
+    {{- toYaml . | nindent 4 }}
+    {{- end }}
+spec:
+  selector:
+    matchLabels:
+      {{- include "matrix-synapse.selectorLabels" . | nindent 6 }}
+      app.kubernetes.io/component: synapse
+  podMetricsEndpoints:
+    - port: metrics
+      path: "/_synapse/metrics"
+{{- end }}

matrix-synapse/templates/prometheus-rules.yaml

@@ -0,0 +1,97 @@
+{{- if and ( .Values.prometheus.rules.enabled ) ( .Capabilities.APIVersions.Has "monitoring.coreos.com/v1" ) }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ include "matrix-synapse.fullname" . }}
+  {{- with .Values.prometheus.rules.namespace }}
+  namespace: {{ . | quote }}
+  {{- end }}
+  labels:
+    {{- include "matrix-synapse.labels" . | nindent 4 }}
+    {{- with .Values.prometheus.rules.labels }}
+    {{- toYaml . | nindent 4 }}
+    {{- end }}
+spec:
+  groups:
+    {{- if .Values.prometheus.rules.defaults.enabled }}
+    - name: {{ template "matrix-synapse.name" . }}-Defaults
+      rules:
+
+        ###
+        ### Prometheus Console Only
+        ### The following rules are only needed if you use the Prometheus Console
+        ### in contrib/prometheus/consoles/synapse.html
+        ###
+        - record: 'synapse_federation_client_sent'
+          labels:
+            type: "EDU"
+          expr: 'synapse_federation_client_sent_edus_total + 0'
+        - record: 'synapse_federation_client_sent'
+          labels:
+            type: "PDU"
+          expr: 'synapse_federation_client_sent_pdu_destinations_count_total + 0'
+        - record: 'synapse_federation_client_sent'
+          labels:
+            type: "Query"
+          expr: 'sum(synapse_federation_client_sent_queries) by (job)'
+
+        - record: 'synapse_federation_server_received'
+          labels:
+            type: "EDU"
+          expr: 'synapse_federation_server_received_edus_total + 0'
+        - record: 'synapse_federation_server_received'
+          labels:
+            type: "PDU"
+          expr: 'synapse_federation_server_received_pdus_total + 0'
+        - record: 'synapse_federation_server_received'
+          labels:
+            type: "Query"
+          expr: 'sum(synapse_federation_server_received_queries) by (job)'
+
+        - record: 'synapse_federation_transaction_queue_pending'
+          labels:
+            type: "EDU"
+          expr: 'synapse_federation_transaction_queue_pending_edus + 0'
+        - record: 'synapse_federation_transaction_queue_pending'
+          labels:
+            type: "PDU"
+          expr: 'synapse_federation_transaction_queue_pending_pdus + 0'
+        ###
+        ### End of 'Prometheus Console Only' rules block
+        ###
+
+
+        ###
+        ### Grafana Only
+        ### The following rules are only needed if you use the Grafana dashboard
+        ### in contrib/grafana/synapse.json
+        ###
+        - record: synapse_storage_events_persisted_by_source_type
+          expr: sum without(type, origin_type, origin_entity) (synapse_storage_events_persisted_events_sep_total{origin_type="remote"})
+          labels:
+            type: remote
+        - record: synapse_storage_events_persisted_by_source_type
+          expr: sum without(type, origin_type, origin_entity) (synapse_storage_events_persisted_events_sep_total{origin_entity="*client*",origin_type="local"})
+          labels:
+            type: local
+        - record: synapse_storage_events_persisted_by_source_type
+          expr: sum without(type, origin_type, origin_entity) (synapse_storage_events_persisted_events_sep_total{origin_entity!="*client*",origin_type="local"})
+          labels:
+            type: bridges
+
+        - record: synapse_storage_events_persisted_by_event_type
+          expr: sum without(origin_entity, origin_type) (synapse_storage_events_persisted_events_sep_total)
+
+        - record: synapse_storage_events_persisted_by_origin
+          expr: sum without(type) (synapse_storage_events_persisted_events_sep_total)
+        ###
+        ### End of 'Grafana Only' rules block
+        ###
+    {{- end }}{{/* end-if .defaults.enabled */}}
+
+    {{- with .Values.prometheus.rules.additionalRules }}
+    - name: {{ template "matrix-synapse.name" $ }}-Additional
+      rules:
+        {{- toYaml . | nindent 4 }}
+    {{- end }}{{/* end-with additionalRules */}}
+{{- end }}

matrix-synapse/values.yaml

@@ -911,3 +911,21 @@ serviceAccount:
   annotations: {}
     # eks.amazonaws.com/role-arn: arn:aws:iam::000000000000:role/matrix-synapse
   # name: non-default-service-name
+
+prometheus:
+  podmonitor:
+    enabled: false
+    labels: {}
+  rules:
+    enabled: false
+    labels: {}
+    defaults:
+      enabled: true
+    additionalRules: []
+
+grafana:
+  dashboards:
+    enabled: false
+    labels:
+      grafana_dashboard: "1"
+    annotations: {}