Benthos is a high performance and resilient stream processor, able to connect various sources and sinks in a range of brokering patterns and perform hydration, enrichments, transformations and filters on payloads.
It comes with a powerful mapping language, is easy to deploy and monitor, and ready to drop into your pipeline either as a static binary, docker image, or serverless function, making it cloud native as heck.
Benthos is fully declarative, with stream pipelines defined in a single config file, allowing you to specify connectors and a list of processing stages:
input: gcp_pubsub: project: foo subscription: bar pipeline: processors: - bloblang: | root.message = this root.meta.link_count = this.links.length() root.user.age = this.user.age.number() output: redis_streams: url: tcp://TODO:6379 stream: baz max_in_flight: 20
Yep, we got 'em. Benthos implements transaction based resiliency with back pressure. When connecting to at-least-once sources and sinks it guarantees at-least-once delivery without needing to persist messages during transit.
Apache Pulsar, AWS (DynamoDB, Kinesis, S3, SQS, SNS), Azure (Blob storage, Queue storage, Table storage), Cassandra, Elasticsearch, File, GCP (Pub/Sub, Cloud storage), HDFS, HTTP (server and client, including websockets), Kafka, Memcached, MQTT, Nanomsg, NATS, NATS JetStream, NATS Streaming, NSQ, AMQP 0.91 (RabbitMQ), AMQP 1, Redis (streams, list, pubsub, hashes), MongoDB, SQL (MySQL, PostgreSQL, Clickhouse, MSSQL), Stdin/Stdout, TCP & UDP, sockets and ZMQ4.
Connectors are being added constantly, if something you want is missing then open an issue.
If you want to dive fully into Benthos then don't waste your time in this dump, check out the documentation site.
For guidance on how to configure more advanced stream processing concepts such as stream joins, enrichment workflows, etc, check out the cookbooks section.
For guidance on building your own custom plugins in Go check out the public APIs.
Grab a binary for your OS from here. Or use this script:
curl -Lsf https://sh.benthos.dev | bash
Or pull the docker image:
docker pull jeffail/benthos
Benthos can also be installed via Homebrew:
brew install benthos
For more information check out the getting started guide.
benthos -c ./config.yaml
Or, with docker:
# Using a config file docker run --rm -v /path/to/your/config.yaml:/benthos.yaml jeffail/benthos # Using a series of -s flags docker run --rm -p 4195:4195 jeffail/benthos \ -s "input.type=http_server" \ -s "output.type=kafka" \ -s "output.kafka.addresses=kafka-server:9092" \ -s "output.kafka.topic=benthos_topic"
Benthos serves two HTTP endpoints for health checks:
/pingcan be used as a liveness probe as it always returns a 200.
/readycan be used as a readiness probe as it serves a 200 only when both the input and output are connected, otherwise a 503 is returned.
Benthos exposes lots of metrics either to Statsd, Prometheus or for debugging purposes an HTTP endpoint that returns a JSON formatted object.
Benthos provides lots of tools for making configuration discovery, debugging and organisation easy. You can read about them here.
Build with Go (1.16 or later):
git clone [email protected]:Jeffail/benthos cd benthos make
Benthos uses golangci-lint for linting, which you can install with:
curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(go env GOPATH)/bin
And then run it with
It's pretty easy to write your own custom plugins for Benthos in Go, for information check out the API docs, and for inspiration there's an example repo demonstrating a variety of plugin implementations.
There's a multi-stage
Dockerfile for creating a Benthos docker image which results in a minimal image from scratch. You can build it with:
Then use the image:
docker run --rm \ -v /path/to/your/benthos.yaml:/config.yaml \ -v /tmp/data:/data \ -p 4195:4195 \ benthos -c /config.yaml
Benthos supports ZMQ4 for both data input and output. To add this you need to install libzmq4 and use the compile time flag when building Benthos:
Or to build a docker image using CGO, which includes ZMQ: