AWS CloudWatch integration

Monitoring: metrics, alarms, dashboards, statistics, anomaly detection

Help CentreConnectors

Overview

Amazon CloudWatch is AWS's native monitoring service – metrics, alarms, dashboards, anomaly detection. The Automize AWS CloudWatch connector covers metric publish + query, alarm CRUD, dashboard management, and anomaly detection model listing.

  • Regions: CloudWatch is regional. Metrics are region-scoped; alarms reference metrics in their own region.
  • Plans: Pay-per-metric + alarm + dashboard. Free tier covers basic AWS service metrics.

Setting up the connection

Auth type: aws_sigv4.

  1. AWS Console → IAM → user with programmatic access.
  2. Attach CloudWatchFullAccess (or scoped subset).
  3. In Automize, open Settings → Connectors → AWS CloudWatch → Add connection. Paste keys + region.

Find your credentials at https://us-east-1.console.aws.amazon.com/iam/home.


Rate limits

Per-region API limits – generous for normal usage. PutMetricData scales horizontally. Connector retries Throttling errors.


Data model

CloudWatch primitives:

    Namespace ─── Metrics (dimension-keyed time series)
    Alarm ── tracks a metric or metric-math expression, triggers actions
    Dashboard ── visualisation composite
    Anomaly Detector ── ML model for a metric, used by alarms

Available operations

16 operations available. Click any row to jump to its detail.

OperationCategoryWhat it does
Delete AlarmsGeneralDelete one or more alarms
Delete DashboardsGeneralDelete one or more dashboards
Describe Alarm HistoryGeneralGet the history of an alarm
Describe AlarmsGeneralList CloudWatch alarms
Describe Anomaly DetectorsGeneralList anomaly detection models
Disable Alarm ActionsGeneralDisable actions for one or more alarms
Enable Alarm ActionsGeneralEnable actions for one or more alarms
Get DashboardGeneralGet the body of a dashboard
Get Metric DataGeneralRetrieve metric data using metric math expressions
Get Metric StatisticsGeneralGet statistics for a metric (avg, sum, min, max, count)
List DashboardsGeneralList CloudWatch dashboards
List MetricsGeneralList available metrics
Put DashboardGeneralCreate or update a dashboard
Put Metric AlarmGeneralCreate or update a metric alarm
Put Metric DataGeneralPublish metric data points
Set Alarm StateGeneralManually set the state of an alarm

Operations

Delete Alarms

Delete one or more alarms

No input parameters.

Tips

  • AlarmNames[] – bulk delete up to 100 per call.

Delete Dashboards

Delete one or more dashboards

No input parameters.

Tips

  • DashboardNames[] – bulk delete.

Describe Alarm History

Get the history of an alarm

No input parameters.

Tips

  • State changes over time. Filter by HistoryItemType (StateUpdate, Action, ConfigurationUpdate).

Describe Alarms

List CloudWatch alarms

No input parameters.

Tips

  • Filter by StateValue (OK, ALARM, INSUFFICIENT_DATA), AlarmNamePrefix.

Describe Anomaly Detectors

List anomaly detection models

No input parameters.

Tips

  • Anomaly detection models attached to metrics. Used by alarms with ThresholdMetricId.

Disable Alarm Actions

Disable actions for one or more alarms

No input parameters.

Tips

  • Mute notifications without deleting alarms. Use for planned downtimes.

Enable Alarm Actions

Enable actions for one or more alarms

No input parameters.

Tips

  • Re-enables alarm notifications after maintenance.

Get Dashboard

Get the body of a dashboard

No input parameters.

Tips

  • Returns dashboard body as JSON (widget definitions).

Get Metric Data

Retrieve metric data using metric math expressions

No input parameters.

Tips

  • Modern API – metric math expressions, multiple metrics in one call.
  • MetricDataQueries[]: id + (MetricStat OR Expression) + Period.

Get Metric Statistics

Get statistics for a metric (avg, sum, min, max, count)

No input parameters.

Tips

  • Legacy – single metric, single aggregation. Prefer Get Metric Data for new builds.

List Dashboards

List CloudWatch dashboards

No input parameters.

Tips

  • Filter by DashboardNamePrefix.

List Metrics

List available metrics

No input parameters.

Tips

  • Filter by Namespace, MetricName, Dimensions[]. Pages 500 per call (NextToken).

Put Dashboard

Create or update a dashboard

No input parameters.

Tips

  • Create or update by name. body: JSON with widgets[].

Put Metric Alarm

Create or update a metric alarm

No input parameters.

Tips

  • Create or update alarm. AlarmName + MetricName + Threshold + ComparisonOperator + EvaluationPeriods + Period.
  • AlarmActions: SNS topic ARN for notifications.

Put Metric Data

Publish metric data points

No input parameters.

Tips

  • Publish custom metrics. MetricData[]: each has MetricName + Dimensions + Value (or Values+Counts for compressed).
  • Up to 20 metrics per call. Higher cadence than once per minute requires Storage Resolution=1.

Set Alarm State

Manually set the state of an alarm

No input parameters.

Tips

  • Manually flip alarm state. Useful for testing notifications without waiting for metric thresholds.

FAQ

Standard vs high-resolution metrics?
Standard = 1-min granularity. High-res = down to 1-sec (StorageResolution=1 on put). High-res costs more but allows tighter alarms.
CloudWatch Logs – covered?
Not in this connector. CloudWatch Logs is a separate service. Use aws-logs (not currently exposed) or stream to your own log store.
Why is my alarm not firing?
Check Treat Missing Data setting – 'missing' (default) keeps state INSUFFICIENT_DATA. For zero-rate metrics, set treatMissingData='notBreaching' or 'breaching'.

Related connectors

See it working on your own data

Everything documented here ships with the platform – try the document tools free, or go live in 7 days.