KafkaFlow is a .NET framework to create Kafka based applications, simple to use and extend.
KafkaFlow uses Confluent Kafka Client.
- Multi-threaded consumer with message order guarantee
- Middlewares support for producing and consuming messages
- Support topics with different message types
- Consumers with many topics
- Serializer middleware with ApacheAvro (with Schema Registry), ProtoBuf and Json support
- Compressor middleware (Gzip or implementing
IMessageCompressor
interface) - Graceful shutdown (wait to finish processing to shutdown)
- Store offset when processing ends, avoiding message loss
- Supports .NET Core and .NET Framework
- Can be used with any dependency injection framework (see here)
- Fluent configuration
- Admin Web API that allows pause, resume and restart consumers, change workers count and rewind offsets, all at runtime
.NET Core 2.1 and later using Hosted Service
public static void Main(string[] args)
{
Host
.CreateDefaultBuilder(args)
.ConfigureServices((hostContext, services) =>
{
services.AddKafkaFlowHostedService(kafka => kafka
.UseConsoleLog()
.AddCluster(cluster => cluster
.WithBrokers(new[] { "localhost:9092" })
.AddConsumer(consumer => consumer
.Topic("sample-topic")
.WithGroupId("sample-group")
.WithBufferSize(100)
.WithWorkersCount(10)
.AddMiddlewares(middlewares => middlewares
.AddSerializer<NewtonsoftJsonMessageSerializer>()
.AddTypedHandlers(handlers => handlers
.AddHandler<SampleMessageHandler>())
)
)
.AddProducer("producer-name", producer => producer
.DefaultTopic("sample-topic")
.AddMiddlewares(middlewares => middlewares
.AddSerializer<NewtonsoftJsonMessageSerializer>()
)
)
)
);
})
.Build()
.Run();
}
See the setup page and samples for more details
Read the Contributing guidelines