Skip to main content
AWS Health sends the service events that affect your account (operational issues, scheduled changes, account notifications) to Amazon EventBridge as events with source set to aws.health and detail-type set to AWS Health Event. The Flashduty AWS EventBridge integration handles these events explicitly, so no separate AWS Health integration is needed: create an AWS EventBridge integration in Flashduty and use its push URL as the target of an EventBridge rule.

​
In Flashduty On-call

Get an integration push URL in either of the two ways below. Choose the AWS EventBridge integration type in both, not AWS Health.

​
Use a dedicated integration

  1. In the Flashduty console, go to Channels and open a channel
  2. Go to Settings → Integrations → Dedicated integrations and click Add an integration
  3. Select AWS EventBridge and click Save
  4. Open the generated integration card and copy the Push URL

​
Use a shared integration

  1. In the Flashduty console, go to Integration Center → Alert Events
  2. Select AWS EventBridge and enter an integration name
  3. Configure the default route and select a channel; you can add more rules under Routes after creation
  4. Click Save and copy the generated Push URL

​
Configure in AWS

AWS Health events are forwarded by an EventBridge rule whose target is an API destination that points to the Flashduty push URL (an SNS topic target also works). The steps for both target types are in the AWS EventBridge integration; for AWS Health you only need the event pattern below.
  1. Create a rule in EventBridge with this event pattern:
  1. To narrow the scope, add filters under detail. For example, only issue events for Amazon EC2:
  1. Choose the API destination that points to the Flashduty push URL as the rule target and save the rule
  2. While creating the rule, you can pick the AWS Health sample events in the Test event pattern panel of the EventBridge console to confirm the pattern matches
Per AWS: rules only receive events for your own account, and public (PUBLIC) events can take up to an hour to start arriving after you create a rule. For events aggregated through AWS Organizations, see the AWS guide Aggregating AWS Health events.
Do not filter out statusCode in the event pattern, for example by matching only "statusCode": ["open"]. Otherwise the notification that the event closed never arrives and the alert cannot recover.

​
Regions and backup rules

  • Create a rule in every Region you want to receive events from. Global events (for example IAM-related events) are delivered to US East (N. Virginia), us-east-1
  • In the standard AWS partition, events are also sent to the backup Region US West (Oregon). If you create rules in both the primary and the backup Region, the same event arrives twice; both copies have the same eventArn and affected account, so Flashduty merges them into one alert through the same Alert Key
  • To keep only the primary Region’s events, add the filter "backupEvent": ["false"] (detail.backupEvent) to the backup Region rule

​
Field mapping

​
Severity

EventBridge events carry no common severity field, so every AWS Health event triggers as Warning. To differentiate, rewrite the severity by the event_type_category label in Alert processing, for example issue to Critical and scheduledChange kept at Warning or lowered to Info.

​
Recovery and deduplication

  • AWS Health sends an update with statusCode set to closed when the event ends, and Flashduty recovers the alert with the same Alert Key
  • Scheduled changes (scheduledChange) are sent with status upcoming before they start, so Flashduty triggers alerts ahead of time. If you only care about ongoing incidents, restrict eventTypeCategory to issue in the event pattern
  • Events with detail-type set to AWS Health Abuse Event have no dedicated handling: each event becomes its own alert keyed by the event id, with no recovery. Matching only AWS Health Event in the pattern avoids them
  • Events without detail.eventArn are rejected (HTTP 400)

​
Troubleshooting

  • The rule does not fire: check the event pattern; public events can take up to an hour after the rule is created; make sure rules exist in the event’s Region and in us-east-1 for global events
  • The alert does not recover: confirm the event pattern does not filter out the closed status and that the target has no Input transformer (the full event must be sent)
  • Two alerts for one event: confirm both events have the same detail.eventArn and affected account; if detail-type was rewritten the event cannot be merged as a Health event
  • The API destination call fails: confirm the endpoint is the full push URL and HTTP method is POST
For field details, see the AWS documentation AWS Health events Amazon EventBridge schema and Creating EventBridge rules for AWS Region coverage.