Microsoft Azure Cosmos DB 429 Request Unit Rate Limit Sentinel

id: azure-cosmos-db-429-ru-rate-limit-sentinel
namespace: company.cloud

inputs:
  - id: subscription_id
    type: STRING
    description: Azure Subscription ID.
  - id: resource_group
    type: STRING
    description: Azure Resource Group name.
  - id: cosmos_account_name
    type: STRING
    description: Azure Cosmos DB account name.
  - id: threshold_429_count
    type: INT
    defaults: 10
    description: "Maximum allowable HTTP 429 rate limit exceptions before alerting
      (default 10)."

tasks:
  - id: check_cosmos_rate_limiting
    type: io.kestra.plugin.scripts.shell.Commands
    containerImage: mcr.microsoft.com/azure-cli:latest
    commands:
      - |
        az login --service-principal -u "{{ secret('AZURE_CLIENT_ID') }}" -p "{{ secret('AZURE_CLIENT_SECRET') }}" --tenant "{{ secret('AZURE_TENANT_ID') }}" --output none
        az account set --subscription "{{ inputs.subscription_id }}"

        RESOURCE_ID="/subscriptions/{{ inputs.subscription_id }}/resourceGroups/{{ inputs.resource_group }}/providers/Microsoft.DocumentDB/databaseAccounts/{{ inputs.cosmos_account_name }}"

        az monitor metrics list           --resource "$RESOURCE_ID"           --metric "TotalRequests"           --filter "StatusCode eq '429'"           --interval PT5M           --output json > /tmp/cosmos_telemetry.json

        EXCEPTIONS_429=$(jq '[.value[]?.timeseries[]?.data[]?.total // 0] | add // 0' /tmp/cosmos_telemetry.json)
        echo "::set-output name=rate_limited_count::$EXCEPTIONS_429::"

  - id: evaluate_rate_limit
    type: io.kestra.plugin.core.flow.If
    condition: "{{ outputs.check_cosmos_rate_limiting.vars.rate_limited_count | int
      > inputs.threshold_429_count }}"
    then:
      - id: notify_data_team
        type: io.kestra.plugin.slack.notifications.SlackIncomingWebhook
        url: "{{ secret('SLACK_WEBHOOK') }}"
        payload: |
          {
            "channel": "#cloud-ops",
            "attachments": [
              {
                "color": "danger",
                "title": "🚨 Azure Cosmos DB Rate Limit (HTTP 429) Alert",
                "text": "Cosmos DB Account `{{ inputs.cosmos_account_name }}` experienced *{{ outputs.check_cosmos_rate_limiting.vars.rate_limited_count }} HTTP 429 exceptions* over the last 5 minutes! Allocated Request Units per second (RU/s) have been saturated by application queries.",
                "footer": "Kestra Cloud Sentinel • Microsoft Azure"
              }
            ]
          }

errors:
  - id: alert_pipeline_failure
    type: io.kestra.plugin.slack.notifications.SlackIncomingWebhook
    url: "{{ secret('SLACK_WEBHOOK') }}"
    payload: |
      {
        "channel": "#cloud-ops",
        "attachments": [
          {
            "color": "danger",
            "title": "⚠️ Azure Cosmos DB Rate Limit Sentinel Failure",
            "text": "Failed to query Azure Monitor metrics for Cosmos DB account. Verify service principal credentials.",
            "footer": "Kestra Workflow Failure Sentinel"
          }
        ]
      }

triggers:
  - id: periodic_schedule
    type: io.kestra.plugin.core.trigger.Schedule
    cron: "*/15 * * * *"
    disabled: true