When your application generates events that require serverless processing, you can use EventBridge event streams to deliver those events to a Function Compute function in real time. An event stream connects an event source to Function Compute through a continuous pipeline that supports filtering, transformation, retry policies, and dead-letter queues.
Prerequisites
Before you begin, make sure that you have:
Create an event stream
Log on to the EventBridge console. In the left-side navigation pane, click Event Streams.
In the top navigation bar, select a region, then click Create Event Stream.
On the Create Event Stream page, enter a Task Name and Description.
Configure the event source and filtering rules
In the Source, Filtering, and Transformation steps, configure the event source, filtering rules, and transformation logic, then click Next Step.
NoteFor details on event transformation, see Use Function Compute to perform message cleansing.
Configure the Function Compute sink
In the Sink step, set Service Type to Function Compute, then configure the following parameters:
Parameter
Description
Example
Service Type
The Function Compute service to receive events. Select the service you created in the prerequisites.
test
Function
The function within the selected service that processes incoming events.
test
Version and Alias
The function version to invoke. Select Specified Version or Specified Alias. If you select Specified Version, you must configure the Version parameter. If you select Specified Alias, you must configure the Alias parameter.
Specified Version / LATEST
Invocation Mode
How EventBridge calls the function. Asynchronous: EventBridge delivers the event without waiting for the function to finish. Synchronous: EventBridge waits for the function to return a response.
Asynchronous
Event Format
The structure in which events are delivered to the function. Object: each event is delivered as a single JSON object -- the function receives one event per invocation. ObjectList: events are batched into a JSON array -- the function must iterate over the array to process each event. This parameter is optional. Default: ObjectList.
Object
Event
The transformation to apply before delivering events. Complete Event delivers the full event payload. For custom transformations, see Event transformation.
Complete Event
Configure retry and dead-letter settings
Under Task Property, configure the retry policy and dead-letter queue for the event stream. The retry policy controls how many times EventBridge retries delivery when the function invocation fails. The dead-letter queue captures events that still fail after all retries, so you can inspect and reprocess them later. For the full list of retry and dead-letter options, see Retry policies and dead-letter queues.
Click Save.
Enable the event stream
Go back to the Event Streams page and find the event stream you created. In the Actions column, click Enable. The event stream takes 30 to 60 seconds to start. Track the progress in the Status column.
Verify event delivery
After the event stream is enabled, confirm that your function receives events:
On the Event Streams page, find your event stream and click the function name in the Event Target column.
On the function details page, click the Logs tab, then click Function Logs to view the invocation logs.
If logs appear with event data from the source, the event stream is working correctly. If no logs appear, check the following:
Verify that the event source is generating events.
Confirm that the filtering rules do not exclude the events you expect.
Check the dead-letter queue for failed delivery attempts.
What's next
Event transformation -- Transform event payloads before they reach the function
Retry policies and dead-letter queues -- Configure failure handling for production workloads