Combines Kinesis mhart/kinesalite and DynamoDB mhart/dynalite
Image also contains aws-cli to allow simpler communication with kinesis
Build docker image
docker build -t kinesis \
--build-arg AWS_ACCESS_KEY_ID=${AWS_ACCESS_KEY_ID} \
--build-arg AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY}\
--build-arg AWS_DEFAULT_REGION=${AWS_DEFAULT_REGION}
Run docker
docker run --name kinesis -d kinesis
Kinesis service exposed on port 443. when running docker, you would probably have to map 443 to a different local port (i.e 4567)
DynamoDB service exposed on port 4568
Once docker starts running kinesalite, an initialization script is executed,
the script init_kinesis.sh
allows you to create streams in advance before the service starts running.
#!/usr/bin/env bash
#
## list all streams to create
#
#declare -a streamNames=(
# "stream1"
# "stream2"
# ...
# "streamN"
#)
#
#for streamName in "${streamNames[@]}"
#do
# ./create_stream.sh ${streamName}
#done
#
docker exec -i kinesis ./create_stream.sh streamname_to_create
aws --endpoint-url='https://localhost*' --no-verify-ssl kinesis
val kinesisClient: AmazonKinesis = AmazonKinesisClientBuilder.standard()
.withCredentials(credentialsProvider)
.withEndpointConfiguration(new EndpointConfiguration("localhost:4567", region))
.build()
val dynamoDBClient: AmazonDynamoDB = AmazonDynamoDBClientBuilder.standard()
.withCredentials(credentialsProvider)
.withEndpointConfiguration(new EndpointConfiguration("localhost:4568", region))
.build()
new Worker.Builder()
.recordProcessorFactory(recordProcessorFactory)
.config(libConfig)
.kinesisClient(kinesisClient)
.dynamoDBClient(dynamoDBClient)
.build()
new KinesisProducerConfiguration()
.setRegion(config.region)
.setCredentialsProvider(CredentialsProvider())
.setMaxConnections(24)
.setRequestTimeout(30000)
.setAggregationEnabled(true)
.setRecordMaxBufferedTime(config.recordMaxBufferedTime.toMillis)
.setKinesisEndpoint("localhost")
.setKinesisPort("4567")
.setVerifyCertificate(false)