The KrakenD Playground is a demonstration environment that puts together the necessary pieces to get you started with our API Gateway, using example use-cases.
As KrakenD is an API gateway, we have added surrounding services:
Click to play on Youtube (with subtitles)
The docker-compose.yml includes the following services:
On port 8080
, you have an instance of KrakenD Community Edition with several endpoints. Its configuration is available at config/krakend/krakend.json
, including descriptive @comments
for each endpoint.
On port 8000
, you have a simple fake API that provides raw data to the gateway. You can add or remove data by adding XML, JSON, or RSS files in the data
folder.
It runs in http://localhost:8000
KrakenD can export telemetry to several services; this demonstration has a few examples. After starting the service and calling some endpoints, you will see the activity in Jaeger, Grafana, and Kibana.
Metrics | Logging | Tracing |
---|---|---|
Grafana shows the metrics stored by KrakenD on InfluxDB (credentials: admin/admin) | Kibana shows the logs registered by Logstash and persisted in Elasticsearch | Jaeger shows the traces of the activity between the client and your end services, including times of hops. |
URL: http://localhost:4000 | URL: http://localhost:5601 | URL: http://localhost:16686 |
NOTE: To import a Kibana dashboard with some valuable metrics, run in the console the following command once all has started:
$ make elastic
This consumer of the API gateway is a simple Express JS application that interacts with KrakenD to fetch the data. All code is under images/spa-auth-web/
.
The client is a Single Page Application using Auth0 to generate JWT tokens.
You don't need to install any npm locally; the docker image will download and install the dependencies in the container.
Visit http://localhost:3000
A RabbitMQ instance is ready to accept AMQP messages to be delivered to the gateway.
You can insert messages in the krakend
queue at http://localhost:15672/#/queues/%2F/krakend (credentials: guest
/guest
) and see how the async agent picks them and delivers them.
A simple implementation of a JWT revoker using the KrakenD remote bloomfilter client.
More information about JWT revoking is available at https://www.krakend.io/docs/authorization/revoking-tokens/
It runs on http://localhost:9000
Create a new SPA application in Auth0 and fill the autogenerated values they give you under images/spa-auth-web/auth0-variables.js
var AUTH0_CLIENT_ID='AUTH0_CLIENT_ID';
var AUTH0_DOMAIN='AUTH0_DOMAIN';
var AUTH0_AUDIENCE = 'AUTH0_AUDIENCE';
You must do this before starting the docker-compose. If you have started docker-compose before setting these variables, you must build the image again with docker compose build web
.
To start the stack included in docker-compose
$ make start
To follow the KrakenD logs after the complete stack is up & running
$ make logs
To shut down the complete stack, removing all the volumes
$ make stop
Fire up your browser, curl, postman, httpie, or anything else you like to interact with any published services.
When you change the krakend.json
, the changes are applied automatically.
💡 Bonus track - Flexible configuration |
---|
We've added an example of flexible configuration, so you can view how it works in practice. Apart from the default config file, krakend.json , you will find a krakend-flexible-config.tmpl that includes some code snippets, invokes a template using variables, and show some basic logic (iterations & conditions). When working with the flexible configuration, you can optionally ask KrakenD to save the "compiled" output to a file. We've added a command make compile-flexible-config so you can see quickly and easily how KrakenD builds the final configuration file based on the existing templates.Internally KrakenD's flexible configuration uses Golang templating syntax. |
To add or remove endpoints, edit the file krakend/krakend.json
. The easiest way to do it is by dragging this file to the KrakenD Designer and downloading the edited file. If you move the downloaded file to krakend/krakend.json
the server will apply the changes automatically.
The following endpoints are worth noticing:
Feature | Endpoint | Description |
---|---|---|
Simple request | /public |
A simple request calling an internal service and returning a JSON response |
Data aggregation and manipulation | /git/{user} |
Aggregating multiple calls concurrently and manipulating some data from GitHub API |
GraphQL to REST transformation | /starwars_films/{movie_id} |
Transforming a GraphQL query into a regular REST API endpoint |
Caching backend responses | /market/cached |
Caching a backend response (based on cache headers provided by the backend) |
Concurrent requests | /market/concurrent |
Using concurrent requests to gather data from Coingecko API |
Sequential calls | /sequential |
Using sequential proxy to build a pipe of sequential calls, using values from 1st call response into 2nd call request |
An aggregated requests with a failing backend | /fail |
An example on how the X-KrakenD-complete header works when a backend fails on an aggregated response |
JWT based Authentication | /private/auth0 |
Protects and endpoint validating JWT tokens issued by Auth0 |
You will find more examples with comments in config/krakend/krakend.json
Everyone can get started with KrakenD at this repository.
Try it out! If it doesn't help you, or you think you can add additional endpoints or middleware integrations, please open a pull request! (We may be too used to KrakenD and don't recognize what a starter demo should include.)
Thanks!
If you have any questions or doubts, you can find our support resources at https://www.krakend.io/support/
Interested in a demo or a trial license of KrakenD Enterprise? Write us »
Note: if you're looking for the KrakenD Enterprise Edition Playground, you'll find it here: https://github.com/krakendio/playground-enterprise