patron

Microservice framework following best cloud practices with a focus on productivity.

Project maintained by beatlabs Hosted on GitHub Pages — Theme by mattgraham

Examples

The examples/ folder contains a number of small services which work together to provide an overview of Patron’s clients and components, as well as the built-in tracing and logging. After you start them all, you can see how a request travels through all services by triggering the start_processing.sh script.

Prerequisites

To run these examples on your system, you need the following dependencies installed:

Before all services start we should first start all dependencies (e.g. kafka, postgres, jaeger, rabbitmq, redis, etc…) with docker-compose.

docker-compose up -d

To tear down the above just:

docker-compose down

Next up, we will start several services that connect to each other as a chain using different communication stacks. All patron services start an HTTP component exposing ‘liveness’, ‘readiness’, ‘metrics’ and ‘debugging’ endpoints, therefore, all the following ‘microservices’ use different ports in order to avoid collisions. Below you can find a simplified view of the communication between the services:

Flow of requests/events

HTTP Cache Service

The service shows how to use:

The service can be started as follows:

go run examples/http-cache/main.go

HTTP Service

The service shows how to use:

The service can be started as follows:

go run examples/http/main.go

HTTP API Key Service

The service can be started as follows:

go run examples/http-sec/main.go

Kafka Service

NOTE

Use either this service or the Legacy Kafka Service. They collide in the served port configuration and therefore cannot coexist.

The service shows how to use:

The service can be started as follows:

go run examples/kafka/main.go

Legacy Kafka Service

NOTE

Use either this service or the Kafka Service. They collide in the served port configuration and therefore cannot coexist.

The service shows how to use:

The service can be started as follows:

go run examples/kafka-legacy/main.go

AMQP Service

The service shows how to use:

The service can be started as follows:

go run examples/amqp/main.go

AWS SQS Service

The service shows how to use:

The service can be started as follows:

go run examples/sqs/main.go

AWS SQS Concurrent Service

The service shows how to use:

The service can be started as follows:

go run examples/sqs-simple/main.go

gRPC Service

The service shows how to use:

The service can be started as follows:

go run examples/grpc/main.go

All the above working together

After all services have been started successfully we can send a request and see how it travels through all of them by running.

../examples/start_processing.sh

After that head over to jaeger and prometheus.

Taking a shortcut:

You can see that the examples folder contains also a Makefile. Instead of executing all the above-mentioned commands yourself you could also simply change your directory to the examples folder and run:

$ make

By doing that you will start the infrastructure containers, start the example services with their logs redirected to examples/tmp/log (there you can check if any of the services fail to start).

$ ls -1 examples/tmp/log/*
examples/tmp/log/http-amqp-svc.log
examples/tmp/log/http-cache-svc.log
examples/tmp/log/http-grpc-svc.log
examples/tmp/log/http-kafka-svc.log
examples/tmp/log/http-sec-svc.log
examples/tmp/log/http-sqs-svc.log
examples/tmp/log/http-svc.log

You will also find a file under examples/tmp/pid.txt which is a repository of the pids of your services:

$ cat examples/tmp/pid.txt 
27733|http-cache-svc
28920|http-svc
29738|http-sec-svc
29875|http-kafka-svc
30047|http-amqp-svc
30191|http-sqs-svc
30315|http-grpc-svc

Compression Middleware

The compression-middleware example showcases the compression middleware with a /foo route that returns some random data.

$ go run examples/compression-middleware/main.go 
$ curl -s localhost:50000/foo | wc -c
1398106
$ curl -s localhost:50000/foo -H "Accept-Encoding: nonexisting" | wc -c
1398106
$ curl -s localhost:50000/foo -H "Accept-Encoding: gzip" | wc -c
1053068
$ curl -s localhost:50000/foo -H "Accept-Encoding: deflate" | wc -c
1053045

It also contains a /hello route used by the next example

Client Decompression

After launching the compression-middleware example, you can run the following to validate that Patron’s HTTP client handles compressed requests transparently.

It creates three requests (with and without an ‘Accept-Encoding’ header), where you can see that the response from the previous example is decompressed automatically.

go run examples/client-decompression/main.go