Add Node Logs feature, which currently only handles journal logs

.yamllint.yml

@@ -10,6 +10,8 @@ ignore:
   - node_modules
   - charts/k8s-monitoring/vendir.lock.yml
   - charts/k8s-monitoring/docs/examples/**/output.yaml
+  - charts/k8s-monitoring/tests/integration/**/output.yaml
+  - charts/k8s-monitoring/tests/platform/**/output.yaml
   - charts/k8s-monitoring-v1/docs/examples/**/output.yaml
   - charts/**/templates
 

charts/feature-node-logs/.helmignore

@@ -0,0 +1,6 @@
+docs
+schema-mods
+tests
+Makefile
+README.md
+README.md.gotmpl

charts/feature-node-logs/Chart.lock

@@ -0,0 +1,3 @@
+dependencies: []
+digest: sha256:643d5437104296e21d906ecb15b2c96ad278f20cfc4af53b12bb6069bd853726
+generated: "2024-08-28T15:09:37.347011-05:00"

charts/feature-node-logs/Chart.yaml

@@ -0,0 +1,13 @@
+---
+apiVersion: v2
+name: k8s-monitoring-feature-node-logs
+description: Kubernetes Observability feature for gathering Cluster Node logs.
+type: application
+sources:
+  - https://github.com/grafana/k8s-monitoring-helm/tree/main/charts/feature-node-logs
+version: 1.0.0
+appVersion: 1.0.0
+maintainers:
+  - email: pete.wall@grafana.com
+    name: petewall
+dependencies: []

charts/feature-node-logs/Makefile

@@ -0,0 +1,38 @@
+HAS_HELM_DOCS := $(shell command -v helm-docs;)
+HAS_HELM_UNITTEST := $(shell helm plugin list | grep unittest 2> /dev/null)
+UPDATECLI_FILES := $(shell yq -e '.dependencies[] | select(.repository == "http*") | ".updatecli-" + .name + ".yaml"' Chart.yaml 2>/dev/null | sort | uniq)
+
+.SECONDEXPANSION:
+README.md: values.yaml Chart.yaml $$(wildcard README.md.gotmpl)
+ifdef HAS_HELM_DOCS
+	helm-docs
+else
+	docker run --rm --volume "$(shell pwd):/helm-docs" -u $(shell id -u) jnorwood/helm-docs:latest
+endif
+
+Chart.lock: Chart.yaml
+	helm dependency update .
+	@touch Chart.lock # Ensure the timestamp is updated
+
+values.schema.json: values.yaml $$(wildcard schema-mods/*)
+	../../scripts/schema-gen.sh .
+
+.updatecli-%.yaml: Chart.yaml
+	../../scripts/charts-to-updatecli.sh Chart.yaml
+
+.PHONY: clean
+clean:
+	rm -f README.md values.schema.json $(UPDATECLI_FILES)
+
+.PHONY: build
+build: README.md Chart.lock values.schema.json $(UPDATECLI_FILES)
+
+.PHONY: test
+test: build
+	helm lint .
+	ct lint --lint-conf ../../.configs/lintconf.yaml --helm-dependency-extra-args=--skip-refresh --charts .
+ifdef HAS_HELM_UNITTEST
+	helm unittest .
+else
+	docker run --rm --volume $(shell pwd):/apps helmunittest/helm-unittest .
+endif

charts/feature-node-logs/README.md

@@ -0,0 +1,56 @@
+<!--
+(NOTE: Do not edit README.md directly. It is a generated file!)
+(      To make changes, please modify README.md.gotmpl and run `helm-docs`)
+-->
+
+# k8s-monitoring-feature-node-logs
+
+![Version: 1.0.0](https://img.shields.io/badge/Version-1.0.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 1.0.0](https://img.shields.io/badge/AppVersion-1.0.0-informational?style=flat-square)
+Kubernetes Observability feature for gathering Cluster Node logs.
+
+The Node Logs feature enables the collection of logs from Kubernetes Cluster Nodes.
+
+## Testing
+
+This chart contains unit tests to verify the generated configuration. A hidden value, `deployAsConfigMap`, will render
+the generated configuration into a ConfigMap object. This ConfigMap is not used during regular operation, but it is
+useful for showing the outcome of a given values file.
+
+The unit tests use this to create an object with the configuration that can be asserted against. To run the tests, use
+`helm test`.
+
+Actual integration testing in a live environment should be done in the main [k8s-monitoring](../k8s-monitoring) chart.
+
+## Maintainers
+
+| Name | Email | Url |
+| ---- | ------ | --- |
+| petewall | <pete.wall@grafana.com> |  |
+<!-- markdownlint-disable no-bare-urls -->
+<!-- markdownlint-disable list-marker-space -->
+## Source Code
+
+* <https://github.com/grafana/k8s-monitoring-helm/tree/main/charts/feature-node-logs>
+<!-- markdownlint-enable list-marker-space -->
+<!-- markdownlint-enable no-bare-urls -->
+
+## Values
+
+### General settings
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| fullnameOverride | string | `""` | Full name override |
+| nameOverride | string | `""` | Name override |
+
+### Journal Logs
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| journal.extraDiscoveryRules | string | `""` | Rule blocks to be added used with the loki.source.journal component for journal logs. These relabeling rules are applied pre-scrape against the targets from service discovery. Before the scrape, any remaining target labels that start with `__` (i.e. `__meta_kubernetes*`) are dropped. ([docs](https://grafana.com/docs/alloy/latest/reference/components/discovery/discovery.relabel/#rule-block)) **Note:** Many field names from journald start with an `_`, such as `_systemd_unit`. The final internal label name would be `__journal__systemd_unit`, with two underscores between `__journal` and `systemd_unit`. |
+| journal.extraLogProcessingBlocks | string | `""` | Stage blocks to be added to the loki.process component for journal logs. ([docs](https://grafana.com/docs/alloy/latest/reference/components/loki/loki.process/#blocks)) This value is templated so that you can refer to other values from this file. |
+| journal.formatAsJson | bool | `false` | Whether to forward the original journal entry as JSON. |
+| journal.jobLabel | string | `"integrations/kubernetes/journal"` | The value for the job label for journal logs. |
+| journal.maxAge | string | `"8h"` | The path to the journal logs on the worker node. |
+| journal.path | string | `"/var/log/journal"` | The path to the journal logs on the worker node. |
+| journal.units | list | `[]` | The list of systemd units to keep scraped logs from. If empty, all units are scraped. |

charts/feature-node-logs/README.md.gotmpl

@@ -0,0 +1,32 @@
+<!--
+(NOTE: Do not edit README.md directly. It is a generated file!)
+(      To make changes, please modify README.md.gotmpl and run `helm-docs`)
+-->
+
+{{ template "chart.header" . }}
+{{ template "chart.deprecationWarning" . }}
+{{ template "chart.badgesSection" . }}
+{{ template "chart.description" . }}
+{{ template "chart.homepageLine" . }}
+
+The Node Logs feature enables the collection of logs from Kubernetes Cluster Nodes.
+
+## Testing
+
+This chart contains unit tests to verify the generated configuration. A hidden value, `deployAsConfigMap`, will render
+the generated configuration into a ConfigMap object. This ConfigMap is not used during regular operation, but it is
+useful for showing the outcome of a given values file.
+
+The unit tests use this to create an object with the configuration that can be asserted against. To run the tests, use
+`helm test`.
+
+Actual integration testing in a live environment should be done in the main [k8s-monitoring](../k8s-monitoring) chart.
+
+{{ template "chart.maintainersSection" . }}
+<!-- markdownlint-disable no-bare-urls -->
+<!-- markdownlint-disable list-marker-space -->
+{{ template "chart.sourcesSection" . }}
+<!-- markdownlint-enable list-marker-space -->
+<!-- markdownlint-enable no-bare-urls -->
+{{ template "chart.requirementsSection" . }}
+{{ template "chart.valuesSection" . }}

charts/feature-node-logs/templates/_collector_validation.tpl

@@ -0,0 +1,13 @@
+{{/* Validates that the Alloy instance is appropriate for the given Node Logs settings */}}
+{{/* Inputs: Values (Node Logs values), Collector (Alloy values), CollectorName (string) */}}
+{{- define "feature.nodeLogs.collector.validate" -}}
+{{- if not (eq .Collector.controller.type "daemonset") }}
+  {{- fail (printf "Node Logs feature requires Alloy to be a DaemonSet.\nPlease set:\n%s:\n  controller:\n    type: daemonset" .CollectorName) }}
+{{- end -}}
+{{- if and (hasPrefix "/var/log" .Values.journal.path) (not .Collector.alloy.mounts.varlog) }}
+  {{- fail (printf "Node Logs feature requires Alloy to mount /var/log.\nPlease set:\n%s:\n  alloy:\n    mounts:\n      varlog: true" .CollectorName) }}
+{{- end -}}
+{{- if .Collector.alloy.clustering.enabled }}
+  {{- fail (printf "Node Logs feature requires Alloy to not be in clustering mode.\nPlease set:\n%s:\n  alloy:\n    clustering:\n      enabled: true" .CollectorName) }}
+{{- end -}}
+{{- end -}}

charts/feature-node-logs/templates/_helpers.tpl

@@ -0,0 +1,17 @@
+{{/*
+Create a default fully qualified name.
+We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec).
+If release name contains chart name it will be used as a full name.
+*/}}
+{{- define "feature.nodeLogs.fullname" -}}
+{{- if .Values.fullnameOverride }}
+{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" | lower }}
+{{- else }}
+{{- $name := default .Chart.Name .Values.nameOverride | lower }}
+{{- if contains $name .Release.Name }}
+{{- .Release.Name | trunc 63 | trimSuffix "-" | lower }}
+{{- else }}
+{{- printf "%s-%s" .Release.Name $name | trunc 63 | trimSuffix "-" | lower }}
+{{- end }}
+{{- end }}
+{{- end }}

charts/feature-node-logs/templates/_module.alloy.tpl

@@ -0,0 +1,49 @@
+{{- define "feature.nodeLogs.module" }}
+declare "node_logs" {
+  argument "logs_destinations" {
+    comment = "Must be a list of log destinations where collected logs should be forwarded to"
+  }
+
+  loki.relabel "journal" {
+    {{- if len .Values.journal.units }}
+    rule {
+      action = "keep"
+      source_labels = ["__journal__systemd_unit"]
+      regex = "{{ join "|" .Values.journal.units }}"
+    }
+    {{- end }}
+    rule {
+      action = "replace"
+      source_labels = ["__journal__systemd_unit"]
+      replacement = "$1"
+      target_label = "unit"
+    }
+  {{- if .Values.journal.extraDiscoveryRules }}
+  {{ .Values.journal.extraDiscoveryRules | indent 2 }}
+  {{- end }}
+
+    forward_to = [] // No forward_to is used in this component, the defined rules are used in the loki.source.journal component
+  }
+
+  loki.source.journal "worker" {
+    path = {{ .Values.journal.path | quote }}
+    format_as_json = {{ .Values.journal.formatAsJson }}
+    max_age = {{ .Values.journal.maxAge | quote }}
+    relabel_rules = loki.relabel.journal.rules
+    labels = {
+      job = {{ .Values.journal.jobLabel | quote }},
+      instance = env("HOSTNAME"),
+    }
+    forward_to = [loki.process.journal_logs.receiver]
+  }
+
+  loki.process "journal_logs" {
+  {{- if .Values.journal.extraLogProcessingBlocks }}
+  {{ tpl .Values.journal.extraLogProcessingBlocks . | indent 2 }}
+  {{ end }}
+    forward_to = argument.logs_destinations.value
+  }
+}
+{{- end -}}
+
+{{- define "feature.nodeLogs.alloyModules" }}{{- end }}

charts/feature-node-logs/templates/_notes.tpl

@@ -0,0 +1,11 @@
+{{- define "feature.nodeLogs.notes.deployments" }}{{- end }}
+
+{{- define "feature.nodeLogs.notes.task" }}
+Gather logs from Kubernetes Nodes
+{{- end }}
+
+{{- define "feature.nodeLogs.notes.actions" }}{{- end }}
+
+{{- define "feature.nodeLogs.summary" -}}
+version: {{ .Chart.Version }}
+{{- end }}

charts/feature-node-logs/templates/configmap.yaml

@@ -0,0 +1,11 @@
+{{- if .Values.deployAsConfigMap }}
+---
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ include "feature.nodeLogs.fullname" . }}
+  namespace: {{ .Release.Namespace }}
+data:
+  module.alloy: |-
+    {{- include "feature.nodeLogs.module" . | indent 4 }}
+{{- end }}

charts/feature-node-logs/tests/default_test.yaml

@@ -0,0 +1,46 @@
+# yamllint disable rule:document-start rule:line-length rule:trailing-spaces
+suite: Test default values
+templates:
+  - configmap.yaml
+tests:
+  - it: should render the default configuration
+    set:
+      deployAsConfigMap: true
+    asserts:
+      - isKind:
+          of: ConfigMap
+      - equal:
+          path: data["module.alloy"]
+          value: |-
+            declare "node_logs" {
+              argument "logs_destinations" {
+                comment = "Must be a list of log destinations where collected logs should be forwarded to"
+              }
+
+              loki.relabel "journal" {
+                rule {
+                  action = "replace"
+                  source_labels = ["__journal__systemd_unit"]
+                  replacement = "$1"
+                  target_label = "unit"
+                }
+
+                forward_to = [] // No forward_to is used in this component, the defined rules are used in the loki.source.journal component
+              }
+
+              loki.source.journal "worker" {
+                path = "/var/log/journal"
+                format_as_json = false
+                max_age = "8h"
+                relabel_rules = loki.relabel.journal.rules
+                labels = {
+                  job = "integrations/kubernetes/journal",
+                  instance = env("HOSTNAME"),
+                }
+                forward_to = [loki.process.journal_logs.receiver]
+              }
+
+              loki.process "journal_logs" {
+                forward_to = argument.logs_destinations.value
+              }
+            }

charts/feature-node-logs/values.schema.json

@@ -0,0 +1,41 @@
+{
+    "$schema": "http://json-schema.org/schema#",
+    "type": "object",
+    "properties": {
+        "deployAsConfigMap": {
+            "type": "boolean"
+        },
+        "fullnameOverride": {
+            "type": "string"
+        },
+        "journal": {
+            "type": "object",
+            "properties": {
+                "extraDiscoveryRules": {
+                    "type": "string"
+                },
+                "extraLogProcessingBlocks": {
+                    "type": "string"
+                },
+                "formatAsJson": {
+                    "type": "boolean"
+                },
+                "jobLabel": {
+                    "type": "string"
+                },
+                "maxAge": {
+                    "type": "string"
+                },
+                "path": {
+                    "type": "string"
+                },
+                "units": {
+                    "type": "array"
+                }
+            }
+        },
+        "nameOverride": {
+            "type": "string"
+        }
+    }
+}