Sink configuration

Sink configuration is also of JSON format. Each sink can be identified with a key.

Example with two S3FileSinks

{
    "default": {
        "type": "s3",
        "localFileSink": {
            "type": "local",
            "outputDir": "/mnt/data/suroServer/fileSink1",
            "bucket": "fileSink1"
        }
    },
    "kafkaSink": {
        "type": "Kafka",
        "client.id": "suroserver-default",
        "metadata.broker.list": "kafkabroker-default.cloud.netflix.com:9092",
        "request.required.acks": 1,
        "kafka.metrics": {
            "kafka.metrics.reporters": "com.netflix.nfkafka.metrics.KafkaServoMetricsReporter"
        }
    },
    "anotherFileSink": {
        "type": "s3",
        "localFileSink": {
            "type": "local",
            "outputDir": "/mnt/data/suroServer/fileSink2",
            "bucket": "fileSink2"
        }
    }
}

There should be at least one sink with the 'default' as id. If there's no 'default' sink, suro server would not start, and will throw an error message saying "default sink should be defined". If a data source is not registered in the routing map, all messages with unregistered topic would be sent to the 'default' sink.