ArangoDB Cluster Health and Supervision Sentinel

id: arangodb-cluster-health-and-supervision-sentinel
namespace: company.team

description: |
  Audits ArangoDB distributed multi-model cluster supervision, monitors coordinator and DBServer node status, classifies cluster health, and dispatches incident alerts to Slack.

triggers:
  - id: every_15_minutes
    type: io.kestra.plugin.core.trigger.Schedule
    description: Executes automated ArangoDB cluster supervision and node health
      audits every 15 minutes.
    cron: "*/15 * * * *"
    disabled: true
    inputs:
      arangodb_host: "arangodb-coordinator"
      arangodb_port: 8529
      database: "_system"

inputs:
  - id: arangodb_host
    type: STRING
    displayName: ArangoDB Coordinator Hostname
    description: Hostname or IP address of the ArangoDB coordinator or cluster
      endpoint reachable by the Kestra worker.
    defaults: "arangodb-coordinator"

  - id: arangodb_port
    type: INT
    displayName: ArangoDB HTTP Port
    description: HTTP port of the ArangoDB cluster service (defaults to 8529).
    defaults: 8529

  - id: database
    type: STRING
    displayName: ArangoDB Database Name
    description: Target database context for administrative health checks (defaults
      to _system).
    defaults: "_system"

tasks:
  - id: probe_cluster_health
    type: io.kestra.plugin.core.http.Request
    description: Queries the ArangoDB Agency supervision endpoint to inspect
      real-time health and heartbeat states of all cluster nodes.
    uri: "http://{{ inputs.arangodb_host }}:{{ inputs.arangodb_port
      }}/_admin/cluster/health"
    method: GET
    headers:
      Authorization: "Basic {{ secret('ARANGODB_AUTH_TOKEN') }}"
      Accept: "application/json"
    options:
      allowFailed: true
    allowFailure: true

  - id: probe_server_role
    type: io.kestra.plugin.core.http.Request
    description: Probes the targeted endpoint to verify it operates as an active
      cluster coordinator.
    uri: "http://{{ inputs.arangodb_host }}:{{ inputs.arangodb_port
      }}/_admin/server/role"
    method: GET
    headers:
      Authorization: "Basic {{ secret('ARANGODB_AUTH_TOKEN') }}"
      Accept: "application/json"
    options:
      allowFailed: true
    allowFailure: true

  - id: probe_engine_version
    type: io.kestra.plugin.core.http.Request
    description: Retrieves ArangoDB software version and storage engine details for
      telemetry logging.
    uri: "http://{{ inputs.arangodb_host }}:{{ inputs.arangodb_port }}/_api/version"
    method: GET
    headers:
      Authorization: "Basic {{ secret('ARANGODB_AUTH_TOKEN') }}"
      Accept: "application/json"
    options:
      allowFailed: true
    allowFailure: true

  - id: classify_cluster_status
    type: io.kestra.plugin.core.debug.Return
    description: Evaluates Agency supervision node states to classify cluster
      vitality into HEALTHY, WARNING_DEGRADED_NODE, CRITICAL_NODE_FAILURE, or
      CLUSTER_UNREACHABLE.
    format: |
      {{ (outputs.probe_cluster_health.code is not defined or outputs.probe_cluster_health.code != 200) ? 'CLUSTER_UNREACHABLE' : ( ((outputs.probe_cluster_health.body | json | jq('[.Health | to_entries[] | select(.value.Status == "FAILED")] | length') | first) > 0) ? 'CRITICAL_NODE_FAILURE' : ( ((outputs.probe_cluster_health.body | json | jq('[.Health | to_entries[] | select(.value.Status == "BAD")] | length') | first) > 0) ? 'WARNING_DEGRADED_NODE' : 'HEALTHY' ) ) }}

  - id: route_by_cluster_status
    type: io.kestra.plugin.core.flow.Switch
    description: Routes execution based on the classified cluster supervision state.
    value: "{{ outputs.classify_cluster_status.value }}"
    cases:
      "HEALTHY":
        - id: log_healthy_state
          type: io.kestra.plugin.core.log.Log
          message: "ArangoDB cluster on {{ inputs.arangodb_host }}:{{ inputs.arangodb_port
            }} is HEALTHY. Total active nodes: {{
            outputs.probe_cluster_health.body | json | jq('.Health | length') |
            first }}. Cluster ID: {{ outputs.probe_cluster_health.body | json |
            jq('.ClusterId') | first }}. Version: {{
            outputs.probe_engine_version.body | json | jq('.version') | first
            }}."

      "WARNING_DEGRADED_NODE":
        - id: alert_warning_slack
          type: io.kestra.plugin.slack.notifications.SlackIncomingWebhook
          url: "{{ secret('SLACK_WEBHOOK') }}"
          payload: |
            {{ {"text": "ArangoDB Cluster Sentinel WARNING on " ~ inputs.arangodb_host ~ ":" ~ inputs.arangodb_port ~ ". Detected degraded node(s) missing heartbeats: " ~ (outputs.probe_cluster_health.body | json | jq('[.Health | to_entries[] | select(.value.Status == \"BAD\") | .key + \" (\" + .value.Role + \":\" + .value.Endpoint + \")\"] | join(\", \")') | first) ~ ". Execution: " ~ execution.id ~ "."} | toJson }}

        - id: log_degraded_warning
          type: io.kestra.plugin.core.log.Log
          message: "ArangoDB cluster degraded node warning: One or more nodes missed
            supervision heartbeats."

      "CRITICAL_NODE_FAILURE":
        - id: alert_critical_slack
          type: io.kestra.plugin.slack.notifications.SlackIncomingWebhook
          url: "{{ secret('SLACK_WEBHOOK') }}"
          payload: |
            {{ {"text": "ArangoDB Cluster Sentinel CRITICAL on " ~ inputs.arangodb_host ~ ":" ~ inputs.arangodb_port ~ ". Confirmed failed cluster node(s): " ~ (outputs.probe_cluster_health.body | json | jq('[.Health | to_entries[] | select(.value.Status == \"FAILED\") | .key + \" (\" + .value.Role + \":\" + .value.Endpoint + \")\"] | join(\", \")') | first) ~ ". Immediate intervention required. Execution: " ~ execution.id ~ "."} | toJson }}

      "CLUSTER_UNREACHABLE":
        - id: alert_unreachable_slack
          type: io.kestra.plugin.slack.notifications.SlackIncomingWebhook
          url: "{{ secret('SLACK_WEBHOOK') }}"
          payload: |
            {{ {"text": "ArangoDB Cluster Sentinel CLUSTER_UNREACHABLE. Cannot reach coordinator at http://" ~ inputs.arangodb_host ~ ":" ~ inputs.arangodb_port ~ "/_admin/cluster/health. Cluster endpoint may be offline or partitioned. Execution: " ~ execution.id ~ "."} | toJson }}

  - id: log_audit_summary
    type: io.kestra.plugin.core.output.OutputValues
    description: Exports structured audit attributes and node supervision metrics
      for reporting dashboards.
    values:
      health_status: "{{ outputs.classify_cluster_status.value }}"
      cluster_id: "{{ outputs.probe_cluster_health.body | json | jq('.ClusterId') |
        first ?? 'UNKNOWN' }}"
      total_nodes: "{{ outputs.probe_cluster_health.body | json | jq('.Health |
        length') | first ?? 0 }}"
      failed_nodes_count: "{{ outputs.probe_cluster_health.body | json | jq('[.Health
        | to_entries[] | select(.value.Status == \"FAILED\")] | length') | first
        ?? 0 }}"
      degraded_nodes_count: "{{ outputs.probe_cluster_health.body | json |
        jq('[.Health | to_entries[] | select(.value.Status == \"BAD\")] |
        length') | first ?? 0 }}"

errors:
  - id: alert_flow_error
    type: io.kestra.plugin.slack.notifications.SlackIncomingWebhook
    description: Dispatches a high-priority Slack notification if an unhandled error
      halts sentinel execution.
    url: "{{ secret('SLACK_WEBHOOK') }}"
    payload: |
      {{ {"text": "ArangoDB Cluster Sentinel workflow error during execution " ~ execution.id ~ " in namespace " ~ flow.namespace ~ "."} | toJson }}