Skip to content

Latest commit



90 lines (68 loc) · 3.94 KB

File metadata and controls

90 lines (68 loc) · 3.94 KB

KMS-AAD Build Status

This git repository is the implementation of my blog post at:


As a JAR File

This is a Java based project built using Gradle. To build use the grade wrapper script in the repository root:

[~eyal]$ ./gradlew build

(or gradlew.bat on Windows). The 'build' task will build and test the project binaries.

As a Docker Image

To build a Docker image use the following command:

[~eyal]$ ./gradlew buildDocker

Which will build the image and tag it as eyallupu/aad-sample-webapp. Alternatively a ready made image can be downloaded directly from Docker Hub:

[~eyal]$ docker pull eyallupu/aad-sample-webapp

Starting The Server


The project is using spring-boot, use the following to build and boot a server (but read this section to the end before doing so!):

[~eyal]$ ./gradlew bootRun

In practice the question if the above will boot successfully or not depends on your environment. As the process has to connect with AWS it requires the correct configuration to authenticate with AWS. If the environment is setup for that (for example the credentials are available via AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY or the default credentials file exists) the above will work, otherwise the environment needs to be setup - see for some hints on how to do that.

By default this service will listen on port 8080.

As a Docker Container

For a successful execution as a Docker container the appropriate AWS credentials and region must be provided to the container. One way is using environment variables as demonstrated below but there are better (and more secure) ways (i.e. mapping ~/.aws into your container as a volume). See for more details on AWS clients environment setup.

[~eyal]$ docker run -d  -P --env AWS_ACCESS_KEY_ID=<replace> \
    --env AWS_SECRET_ACCESS_KEY=<replace> --env AWS_REGION=<replace>  \

The next is to note down the host port which was mapped into the container's 8080 port, the following will list that out:

[~eyal]$ docker ps -n 1 --format "{{.ID}}" | xargs docker port
8080/tcp ->

In the above example the port is 32768.

Client Example

Once the server is running we can give it a try using CURL. The server is preconfigured with two users: 'bob' and 'alice' with passwords the same as the user name (e.g. bob/bob). Those users can be used to demo the usage of AWS following those steps:

We can upload a secret as Alice (remember to replace host and port with actual values):

[~eyal]$ curl -X POST -H 'Content-type: text/plain' -d 'my-secret-value' \
   --user alice:alice  http://localhost:8080/secrets/my-secret-name

Next we can retrieve the secret using Alice's credentials:

[~eyal]$ curl -X GET --user alice:alice  http://localhost:8080/secrets/my-secret-name

However, if Bob will try to fetch the same secret he will be blocked by the KMS

[~eyal]$ curl -X GET --user bob:bob  http://localhost:8080/secrets/my-secret-name

{"timestamp":1517092879867,"status":500,"error":"Internal Server Error",
 "message":"org.springframework.web.util.NestedServletException: Request processing failed; nested exception is null (Service: AWSKMS; Status Code: 400; Error Code: InvalidCiphertextException; Request ID: 3a45ce22-03b4-11e8-8872-c1cc2f0068f1)",


Do remember that this example is using AWS KMS which involves costs for both master keys (CMK) and API calls.