Streaming messaging data in Amazon Chime SDK messaging - Amazon Chime SDK

Streaming messaging data in Amazon Chime SDK messaging

You can configure an AppInstance to receive data, such as messages and channel events, in the form of a stream. You can then react to that data in real time. Currently, Amazon Chime SDK messaging only accepts Kinesis streams as stream destinations. You must have these prerequisites to use Kinesis streams with this feature:

  • Kinesis streams must be in the same AWS account as the AppInstance.

  • A stream must be in the same region as the AppInstance.

  • Stream names have a prefix that starts with chime-messaging-.

  • You must configure at least two shards. Each shard can receive data up to 1MB per second, so scale your stream accordingly.

  • You must enable server-side encryption (SSE).

To configure a Kinesis stream

  1. Create one or more Kinesis streams using the prerequisites in the previous section, then get the ARN. Ensure the caller has Kinesis permissions in addition to Amazon Chime permissions.

    The following examples show how to use the AWS CLI to create a Kinesis stream with two shards, and how to enable SSE.

    aws kinesis create-stream --stream-name chime-messaging-unique-name --shard-count 2

    aws kinesis start-stream-encryption --stream-name chime-messaging-unique-name --encryption-type KMS --key-id "alias/aws/kinesis"

  2. Configure streaming by calling the PutMessagingStreamingConfigurations API.

    You can configure one or both of two data types, and you can choose the same stream or separate streams for them.

    The following examples show how to use the AWS CLI to configure an appinstance to stream the ChannelMessage and Channel data types.

    aws chime-sdk-messaging put-messaging-streaming-configurations --app-instance-arn app_instance_arn \
--streaming-configurations DataType=ChannelMessage,ResourceArn=kinesis_data_stream_arn
    aws chime-sdk-messaging put-messaging-streaming-configurations --app-instance-arn app_instance_arn \
--streaming-configurations DataType=Channel,ResourceArn=kinesis_data_stream_arn

    The data types have the following scopes:

    DataType Event types generated

    ChannelMessage

    CREATE_CHANNEL_MESSAGE

    REDACT_CHANNEL_MESSAGE

    UPDATE_CHANNEL_MESSAGE

    DELETE_CHANNEL_MESSAGE

    Channel

    CREATE_CHANNEL

    CREATE_SUB_CHANNEL

    UPDATE_CHANNEL

    DELETE_CHANNEL

    UPDATE_CHANNEL_EXPIRATION_SETTINGS

    DELETE_SUB_CHANNEL

    CREATE_CHANNEL_MEMBERSHIP

    DELETE_CHANNEL_MEMBERSHIP

    CREATE_CHANNEL_BAN

    DELETE_CHANNEL_BAN

    CREATE_CHANNEL_MODERATOR

    DELETE_CHANNEL_MODERATOR

  3. Start reading the data from your configured Kinesis stream.

    Note

    Any events sent before you configure streaming are not sent to your Kinesis stream.

Data format

Kinesis outputs records in JSON format with the following fields: EventType and Payload. The payload format depends on the EventType. The following table lists the event types and their corresponding payload formats.

EventType Payload format

CREATE_CHANNEL_MESSAGE

Channel message

REDACT_CHANNEL_MESSAGE

UPDATE_CHANNEL_MESSAGE

DELETE_CHANNEL_MESSAGE

CREATE_CHANNEL

Channel

UPDATE_CHANNEL

DELETE_CHANNEL

UPDATE_CHANNEL_EXPIRATION_SETTINGS

CREATE_CHANNEL_MEMBERSHIP

ChannelMembership

DELETE_CHANNEL_MEMBERSHIP

CREATE_CHANNEL_BAN

ChannelBan

DELETE_CHANNEL_BAN

CREATE_CHANNEL_MODERATOR

ChannelModerator

DELETE_CHANNEL_MODERATOR

CREATE_SUB_CHANNEL

channelARN

SubChannelId

DELETE_SUB_CHANNEL