[Creator] [design-spec] k8s-openobserve-healthcheck

codebundles/k8s-openobserve-healthcheck/.runwhen/generation-rules/k8s-openobserve-healthcheck.yaml

@@ -0,0 +1,23 @@
+apiVersion: runwhen.com/v1
+kind: GenerationRules
+spec:
+  generationRules:
+    - resourceTypes:
+        - deployment
+        - statefulset
+      matchRules:
+        - type: pattern
+          pattern: "openobserve"
+          properties: [name]
+          mode: substring
+      slxs:
+        - baseName: k8s-openobserve-hc
+          shortenedBaseName: openobs-hc
+          qualifiers: ["resource", "namespace", "cluster"]
+          baseTemplateName: k8s-openobserve-healthcheck
+          levelOfDetail: basic
+          outputItems:
+            - type: slx
+            - type: sli
+            - type: runbook
+              templateName: k8s-openobserve-healthcheck-taskset.yaml

codebundles/k8s-openobserve-healthcheck/.runwhen/templates/k8s-openobserve-healthcheck-sli.yaml

@@ -0,0 +1,63 @@
+apiVersion: runwhen.com/v1
+kind: ServiceLevelIndicator
+metadata:
+  name: {{slx_name}}
+  labels:
+    {% include "common-labels.yaml" %}
+  annotations:
+    {% include "common-annotations.yaml" %}
+    runwhen.com/sli: "true"
+spec:
+  displayUnitsLong: OK
+  displayUnitsShort: ok
+  locations:
+    - {{default_location}}
+  description: Aggregates OpenObserve API reachability, workload replica readiness, and storage health for {{match_resource.resource.metadata.name}}.
+  codeBundle:
+    {% if repo_url %}
+    repoUrl: {{repo_url}}
+    {% else %}
+    repoUrl: https://github.com/runwhen-contrib/rw-cli-codecollection.git
+    {% endif %}
+    {% if ref %}
+    ref: {{ref}}
+    {% else %}
+    ref: main
+    {% endif %}
+    pathToRobot: codebundles/k8s-openobserve-healthcheck/sli.robot
+  intervalStrategy: intermezzo
+  intervalSeconds: 300
+  configProvided:
+    - name: CONTEXT
+      value: "{{context}}"
+    - name: NAMESPACE
+      value: "{{match_resource.resource.metadata.namespace}}"
+    - name: OBJECT_NAME
+      value: "{{match_resource.resource.metadata.name}}"
+    - name: KUBERNETES_DISTRIBUTION_BINARY
+      value: "{{custom.kubernetes_distribution_binary | default('kubectl')}}"
+    - name: OPENOBSERVE_SERVICE_NAME
+      value: ""
+    - name: OPENOBSERVE_HTTP_PORT
+      value: "5080"
+    - name: OPENOBSERVE_LOCAL_PORT
+      value: "5080"
+    - name: OPENOBSERVE_ORGANIZATION
+      value: "default"
+    - name: OPENOBSERVE_BASE_URL
+      value: ""
+    - name: OPENOBSERVE_API_TIMEOUT
+      value: "25"
+    - name: STORAGE_S3_FAILURE_THRESHOLD
+      value: "10"
+  secretsProvided:
+  {% if wb_version %}
+    {% include "kubernetes-auth.yaml" ignore missing %}
+  {% else %}
+    - name: kubeconfig
+      workspaceKey: {{custom.kubeconfig_secret_name | default("kubeconfig")}}
+  {% endif %}
+  alertConfig:
+    tasks:
+      persona: eager-edgar
+      sessionTTL: 10m

codebundles/k8s-openobserve-healthcheck/.runwhen/templates/k8s-openobserve-healthcheck-slx.yaml

@@ -0,0 +1,31 @@
+apiVersion: runwhen.com/v1
+kind: ServiceLevelX
+metadata:
+  name: {{slx_name}}
+  labels:
+    {% include "common-labels.yaml" %}
+  annotations:
+    {% include "common-annotations.yaml" %}
+spec:
+  imageURL: https://storage.googleapis.com/runwhen-nonprod-shared-images/icons/kubernetes/resources/labeled/ns.svg
+  alias: OpenObserve Health Check for {{match_resource.resource.metadata.name}}
+  asMeasuredBy: OpenObserve workload / Pod / API health and storage metrics in Kubernetes.
+  configProvided:
+    - name: OBJECT_NAME
+      value: {{match_resource.resource.metadata.name}}
+  owners:
+    - {{workspace.owner_email}}
+  statement: The OpenObserve deployment in the namespace should have ready replicas, stable Pods, a reachable HTTP API, and healthy storage/ingestion/cluster state.
+  additionalContext:
+    {% include "kubernetes-hierarchy.yaml" ignore missing %}
+    qualified_name: "{{ match_resource.qualified_name }}"
+  tags:
+    {% include "kubernetes-tags.yaml" ignore missing %}
+    - name: cloud
+      value: kubernetes
+    - name: service
+      value: openobserve
+    - name: scope
+      value: project
+    - name: access
+      value: read-only

codebundles/k8s-openobserve-healthcheck/.runwhen/templates/k8s-openobserve-healthcheck-taskset.yaml

@@ -0,0 +1,73 @@
+apiVersion: runwhen.com/v1
+kind: Runbook
+metadata:
+  name: {{slx_name}}
+  labels:
+    {% include "common-labels.yaml" %}
+  annotations:
+    {% include "common-annotations.yaml" %}
+spec:
+  location: {{default_location}}
+  description: OpenObserve workload, Pod, API, storage, cluster, ingestion queue, and PVC health checks for {{match_resource.resource.metadata.name}} in namespace {{match_resource.resource.metadata.namespace}}.
+  codeBundle:
+    {% if repo_url %}
+    repoUrl: {{repo_url}}
+    {% else %}
+    repoUrl: https://github.com/runwhen-contrib/rw-cli-codecollection.git
+    {% endif %}
+    {% if ref %}
+    ref: {{ref}}
+    {% else %}
+    ref: main
+    {% endif %}
+    pathToRobot: codebundles/k8s-openobserve-healthcheck/runbook.robot
+  configProvided:
+    - name: CONTEXT
+      value: "{{context}}"
+    - name: NAMESPACE
+      value: "{{match_resource.resource.metadata.namespace}}"
+    - name: OBJECT_NAME
+      value: "{{match_resource.resource.metadata.name}}"
+    - name: KUBERNETES_DISTRIBUTION_BINARY
+      value: "{{custom.kubernetes_distribution_binary | default('kubectl')}}"
+    - name: OPENOBSERVE_SERVICE_NAME
+      value: ""
+    - name: OPENOBSERVE_HTTP_PORT
+      value: "5080"
+    - name: OPENOBSERVE_LOCAL_PORT
+      value: "5080"
+    - name: OPENOBSERVE_ORGANIZATION
+      value: "default"
+    - name: OPENOBSERVE_BASE_URL
+      value: ""
+    - name: OPENOBSERVE_API_TIMEOUT
+      value: "25"
+    - name: OPENOBSERVE_PF_WAIT_SECS
+      value: "15"
+    - name: OPENOBSERVE_CREDENTIAL_SECRET_NAME
+      value: "{{custom.openobserve_credential_secret_name | default('')}}"
+    - name: OPENOBSERVE_CREDENTIAL_USER_KEY
+      value: ""
+    - name: OPENOBSERVE_CREDENTIAL_PASSWORD_KEY
+      value: ""
+    - name: OPENOBSERVE_ALLOW_EXEC
+      value: "false"
+    - name: OPENOBSERVE_POD_RESTART_THRESHOLD
+      value: "5"
+    - name: STORAGE_S3_FAILURE_THRESHOLD
+      value: "10"
+    - name: INGESTION_QUEUE_KEY_THRESHOLD
+      value: "100000"
+    - name: CLUSTER_PENDING_JOBS_THRESHOLD
+      value: "100"
+    - name: PVC_USAGE_THRESHOLD_PERCENT
+      value: "80"
+  secretsProvided:
+  {% if wb_version %}
+    {% include "kubernetes-auth.yaml" ignore missing %}
+  {% else %}
+    - name: kubeconfig
+      workspaceKey: {{custom.kubeconfig_secret_name | default("kubeconfig")}}
+    - name: openobserve_credentials
+      workspaceKey: {{custom.openobserve_credentials_secret_name | default("openobserve_credentials")}}
+  {% endif %}

codebundles/k8s-openobserve-healthcheck/.test/README.md

@@ -0,0 +1,48 @@
+# k8s-openobserve-healthcheck - Tests
+
+## Overview
+
+This `.test/` directory provides infrastructure and Taskfile automation used to
+validate the `k8s-openobserve-healthcheck` CodeBundle against a Kubernetes
+cluster with RunWhen Local.
+
+## Prerequisites
+
+- A running Kubernetes cluster (local `kind`/`minikube`/k3s or a real cluster).
+- `kubectl` configured with a kubeconfig file.
+- Docker (to run the `runwhen-local` container).
+- `task`, `yq`, `curl`, `ajv` on PATH.
+
+## Quick start
+
+1. Point `kubectl` at your test cluster.
+2. Drop a kubeconfig file named `kubeconfig.secret` in this directory
+   (used by RunWhen Local discovery).
+3. `task build-infra` - creates the test namespace and OpenObserve resources.
+4. `task default` - runs discovery and generates SLX/SLI/runbook configs under
+   `output/workspaces/`.
+5. `task clean` - removes the test objects and discovery output.
+
+## Test resources
+
+The `kubernetes/manifest.yaml` creates a namespace `test-openobserve-health`
+containing:
+
+- a Secret `openobserve-credentials` with `ROOT_USER_EMAIL` /
+  `ROOT_USER_PASSWORD` (matching Helm defaults)
+- a Service `openobserve` exposing port `5080`
+- a Deployment `openobserve` (name matches the `openobserve` substring used by
+  the generation rule)
+- a PVC `openobserve-data`
+
+The generation rule matches the Deployment by name, so discovery should produce
+one SLX named after the workload `openobserve` in `test-openobserve-health`.
+
+## Scenarios
+
+The deployment declares `ZO_LOCAL_MODE=true` and a bound PVC. For failure
+scenarios you can edit the manifest before applying:
+
+- `replicas: 2` with a broken image -> replica/pod health issues.
+- Unbind the PVC (`persistentVolumeReclaimPolicy` / delete the PV) -> PVC issues.
+- Point the Service at a bogus port -> API health issues.