Queries

Views use a powerful query system to aggregate and transform data from multiple sources. You can query config items, changes, and metrics to populate your views.

Query Types

Config Queries

Query config items from your catalog:

deployments.yaml
apiVersion: mission-control.flanksource.com/v1
kind: View
metadata:
  name: deployments
  namespace: mc
spec:
  display:
    title: Helm Releases
    icon: rocket
    sidebar: true
  panels:
    - name: Health
      description: Helm Releases grouped by health
      type: piechart
      piechart:
        showLabels: true
        colors:
          healthy: "#28C19B"
          unhealthy: "#F04E6E"
      query: SELECT COUNT(*) AS count, health FROM helm_releases GROUP BY health
    - name: Status
      description: Helm Releases grouped by status
      type: piechart
      piechart:
        colors:
          InstallSucceeded: "#28C19B"
          RollbackSucceeded: "#F4B23C"
          UpgradeFailed: "#F04E6E"
          UpgradeSucceeded: "#5965F2"
      query: SELECT COUNT(*) AS count, status FROM helm_releases GROUP BY status
  columns:
    - name: id
      type: string
      description: The id of the deployment.
      primaryKey: true
      hidden: true
    - name: url
      type: url
      for: application
    - name: application
      type: string
      description: The application name.
    - name: namespace
      type: string
      description: The namespace name.
    - name: chart
      type: string
      description: The chart name.
    - name: version
      type: string
      description: The version of the deployment.
    - name: status
      type: status
      description: The status of the deployment.
    - name: health
      type: health
      description: The health of the deployment.
    - name: lastUpdated
      type: datetime
      description: The last updated time.
  queries:
    helm_releases:
      configs:
        types:
          - Kubernetes::HelmRelease
  mapping:
    application: row.name
    namespace: row.tags.namespace
    chart: row.config.status.history[0].chartName
    version: row.config.status.history[0].chartVersion
    lastUpdated: row.updated_at
    url: >
      "/catalog/" + row.id

Change Queries

Query configuration changes and audit data:

backups.yaml
apiVersion: mission-control.flanksource.com/v1
kind: View
metadata:
  name: backups
  namespace: mc
spec:
  display:
    title: Backups
    icon: database
    sidebar: true
  cache:
    maxAge: 1h
    minAge: 1m
    refreshTimeout: 10s
  columns:
    - name: id
      type: string
      description: The id of the database.
      primaryKey: true
      hidden: true
    - name: database
      type: string
      description: The name of the helm release.
    - name: date
      type: datetime
      description: The namespace name.
    - name: status
      type: status
      description: The status of the backup.
    - name: source
      type: string
      description: The source of the backup.
  queries:
    backups:
      changes:
        search: change_type=BackupSuccessful
        limit: 10
  mapping:
    database: row.name
    date: row.created_at
    status: row.details.status

Metric Queries

Query time-series data from Prometheus and other sources. Metric queries are not yet implemented in the current version, but the structure would be:

resource-usage.yaml
apiVersion: mission-control.flanksource.com/v1
kind: View
metadata:
  name: resource-usage
  namespace: mc
spec:
  queries:
    cpu:
      prometheus:
        url: https://prometheus.demo.prometheus.io
        query: container_cpu_usage_seconds_total{id="/", job="cadvisor"}
    memory:
      prometheus:
        url: https://prometheus.demo.prometheus.io
        query: container_memory_working_set_bytes{id="/", job="cadvisor"}
  panels:
    - name: cpu
      type: number
      query: SELECT SUM(value) as value FROM cpu
      number:
        unit: seconds
        precision: 2
    - name: memory
      type: number
      query: SELECT SUM(value) as value FROM memory
      number:
        unit: bytes

Multi-Source Queries

Views can combine data from different sources and use SQL queries in panels to aggregate and analyze the data:

pipelines.yaml
apiVersion: mission-control.flanksource.com/v1
kind: View
metadata:
  name: pipelines
  namespace: mc
spec:
  display:
    title: Pipelines
    icon: github
    sidebar: true
  cache:
    maxAge: 30m
    minAge: 1m
    refreshTimeout: 30s
  panels:
    - name: Repository
      type: table
      query: SELECT COUNT(*) AS value, details->'repository'->>'full_name' as repository FROM workflows GROUP BY details->'repository'->>'full_name'
    - name: Average Duration
      description: Create Release average duration
      type: number
      number:
        unit: seconds
      query: >
        SELECT AVG(details->'duration') AS value, details->'repository'->>'full_name' as label
        FROM workflows GROUP BY details->'repository'->>'full_name'
  columns:
    - name: id
      type: string
      description: The id of the pipeline.
      primaryKey: true
      hidden: true
    - name: name
      type: string
      description: The name of the pipeline.
    - name: repository
      type: string
      description: The repository of the pipeline.
    - name: lastRun
      type: datetime
      description: The namespace name.
    - name: lastRunBy
      type: string
      description: The chart name.
    - name: duration
      type: duration
      description: The duration of the pipeline.
    - name: status
      type: status
      description: The status of the pipeline.
  queries:
    workflows:
      changes:
        limit: 10
        search: change_type=GitHubActionRun* @order=-created_at
        types:
          - GitHubAction::Workflow
  mapping:
    lastRun: row.details.created_at
    lastRunBy: row.details.triggering_actor.login
    repository: row.details.repository.full_name
    status: >
      has(row.details.conclusion) ? row.details.conclusion : 'pending'
    duration: timestamp(row.details.updated_at) - timestamp(row.details.run_started_at)

This example shows:

• Multiple query sources: Combining workflow runs from changes
• Panel queries: Using SQL to aggregate repository data
• CEL expressions: Complex duration calculations in mappings
• Data aggregation: COUNT (count) and AVG (average) functions in panel queries