springboot_xray

This repository shows how AWS X-Ray can be integrated with Springboot microservices in order to track requests and trace the application flow not only on the individual service but accross the all microservices involved.

To build:

$ ./gradlew clean build

To generate the docker images:

$ ./gradlew frontend_service:docker persistence_service:docker booking_services:manage_service:docker booking_services:search_service:docker  # (I haven't found a way of ignoring the docker task on the root project)

The solution has the following flow:

                                                _____________________________
 ____________________________________          |                             |          ____________________________
|                                    |         |   Booking services:         |         |                            |
|  Frontend service                  |  ---->  |    ______________________   |  ---->  |  Persistence service       |
|  (request validation & routing)    |         |   | Manage reservations  |  |         |                            |
|____________________________________|         |   |______________________|  |         |____________________________|
                                               |    ______________________   |
                                               |   | Slot search          |  |
                                               |   |______________________|  |
                                               |_____________________________|

Docker-compose is used to deploy all the applications and the necessary resources. Check (the docker-compose file)[docker-compose.yml]. The additional components needed are:

• MySQL
• AWS X-Ray Daemon, which forwards the traces sent by our springboot app to AWS

Before starting the stack, the X-Ray Daemon container expects the credentials to be added on the file aws-credentials.env, with the following content:

AWS_REGION=
AWS_ACCESS_KEY_ID=
AWS_SECRET_ACCESS_KEY=

To start the stack:

$ docker-compose up -d
$ docker-compose logs -f

Now you can invoke the application, notice that each request will return a header called X-Amzn-Trace-Id, this is the trace id we can use to check the invocation flow on AWS X-Ray Console:

$ curl -v -H 'x-someheader: somevalue' http://localhost:8080/reservations/search # search for reservations
> ...
> GET /reservations/search HTTP/1.1
> ...
< ...
< X-Amzn-Trace-Id: Root=1-5fc756bd-43220164e40e28bf33de923b;
< ...
["slot 1","slot 2"]%

$ curl -v -H 'x-someheader: somevalue' http://localhost:8080/reservations -d 'reservation=1234' # create new reservation
> ...
> POST /reservations HTTP/1.1
> ...
< ...
< X-Amzn-Trace-Id: Root=1-5fc756fd-568fd3271a4fd2f971c76387;
< ...
reservation=1234 - done at 1606899453377%

And finally you can see the entire application flow tracing on the X-Ray console: