feat: add support for customizing wal settings (#337)

* feat: add support for customizing wal settings * docs: update examples to include the wal settings customization * docs: update readme values to include the wal settings customization
grafana · Jan 16, 2024 · ad28c9b · ad28c9b
1 parent de79fdc
commit ad28c9b
Show file tree

Hide file tree

Showing 38 changed files with 340 additions and 0 deletions.
diff --git a/charts/k8s-monitoring/README.md b/charts/k8s-monitoring/README.md
@@ -139,6 +139,9 @@ The Prometheus and Loki services may be hosted on the same cluster, or remotely
 | externalServices.prometheus.tenantId | string | `""` | Sets the X-Scope-OrgID header when sending metrics |
 | externalServices.prometheus.tenantIdKey | string | `"tenantId"` | The key for the tenant ID property in the secret |
 | externalServices.prometheus.tls | object | `{}` | TLS setting to configure for the metrics service. For remoteWrite protocol, refer to https://grafana.com/docs/agent/latest/flow/reference/components/prometheus.remote_write/#tls_config-block For otlp protocol, refer to https://grafana.com/docs/agent/latest/flow/reference/components/otelcol.exporter.otlp/#tls-block For otlphttp protocol, refer to https://grafana.com/docs/agent/latest/flow/reference/components/otelcol.exporter.otlphttp/#tls-block |
+| externalServices.prometheus.wal.maxKeepaliveTime | string | `"8h"` | Maximum time to keep data in the WAL before removing it. |
+| externalServices.prometheus.wal.minKeepaliveTime | string | `"5m"` | Minimum time to keep data in the WAL before it can be removed. |
+| externalServices.prometheus.wal.truncateFrequency | string | `"2h"` | How frequently to clean up the WAL. |
 | externalServices.prometheus.writeEndpoint | string | `"/api/prom/push"` | Prometheus metrics write endpoint. Preset for Grafana Cloud Metrics instances. |
 | externalServices.prometheus.writeRelabelConfigRules | string | `""` | Rule blocks to be added to the write_relabel_config block of the prometheus.remote_write component. See https://grafana.com/docs/agent/latest/flow/reference/components/prometheus.remote_write/#write_relabel_config-block |
 | externalServices.tempo.authMode | string | `"basic"` | one of "none", "basic" |

diff --git a/charts/k8s-monitoring/templates/agent_config/_metrics_service_remote_write.river.txt b/charts/k8s-monitoring/templates/agent_config/_metrics_service_remote_write.river.txt
@@ -25,6 +25,12 @@ prometheus.remote_write "metrics_service" {
 {{- end }}
   }
 
+  wal {
+    truncate_frequency = {{ .wal.truncateFrequency | quote }}
+    min_keepalive_time = {{ .wal.minKeepaliveTime | quote }}
+    max_keepalive_time = {{ .wal.maxKeepaliveTime | quote }}
+  }
+
   external_labels = {
   {{- range $key, $value := .externalLabels }}
     {{ $key }} = {{ $value | quote }},

diff --git a/charts/k8s-monitoring/values.schema.json b/charts/k8s-monitoring/values.schema.json
@@ -228,6 +228,20 @@
                         "tls": {
                             "type": "object"
                         },
+                        "wal": {
+                            "type": "object",
+                            "properties": {
+                                "truncateFrequency": {
+                                    "type": "string"
+                                },
+                                "minKeepaliveTime": {
+                                    "type": "string"
+                                },
+                                "maxKeepaliveTime": {
+                                    "type": "string"
+                                }
+                            }
+                        },
                         "writeEndpoint": {
                             "type": "string"
                         },

diff --git a/charts/k8s-monitoring/values.yaml b/charts/k8s-monitoring/values.yaml
@@ -85,6 +85,17 @@ externalServices:
         checkInterval: 1s
         # -- Maximum amount of memory targeted to be allocated by the process heap.
         limit: 0MiB
+
+    # Write-Ahead Log (WAL) settings. Only applies when protocol is "remote_write"
+    wal:
+      # -- How frequently to clean up the WAL.
+      truncateFrequency: 2h
+
+      # -- Minimum time to keep data in the WAL before it can be removed.
+      minKeepaliveTime: 5m
+
+      # -- Maximum time to keep data in the WAL before removing it.
+      maxKeepaliveTime: 8h
 
   # Connection information for Grafana Loki
   loki:

diff --git a/examples/control-plane-metrics/metrics.river b/examples/control-plane-metrics/metrics.river
diff --git a/examples/control-plane-metrics/output.yaml b/examples/control-plane-metrics/output.yaml
diff --git a/examples/custom-allow-lists/metrics.river b/examples/custom-allow-lists/metrics.river
diff --git a/examples/custom-allow-lists/output.yaml b/examples/custom-allow-lists/output.yaml
diff --git a/examples/custom-config/metrics.river b/examples/custom-config/metrics.river
diff --git a/examples/custom-config/output.yaml b/examples/custom-config/output.yaml
diff --git a/examples/default-values/metrics.river b/examples/default-values/metrics.river
diff --git a/examples/default-values/output.yaml b/examples/default-values/output.yaml
diff --git a/examples/eks-fargate/metrics.river b/examples/eks-fargate/metrics.river
diff --git a/examples/eks-fargate/output.yaml b/examples/eks-fargate/output.yaml
diff --git a/examples/extra-rules/metrics.river b/examples/extra-rules/metrics.river
diff --git a/examples/extra-rules/output.yaml b/examples/extra-rules/output.yaml
diff --git a/examples/gke-autopilot/metrics.river b/examples/gke-autopilot/metrics.river
diff --git a/examples/gke-autopilot/output.yaml b/examples/gke-autopilot/output.yaml
diff --git a/examples/ibm-cloud/metrics.river b/examples/ibm-cloud/metrics.river
diff --git a/examples/ibm-cloud/output.yaml b/examples/ibm-cloud/output.yaml
diff --git a/examples/metrics-only/metrics.river b/examples/metrics-only/metrics.river