This source code repository stores the configurations to create a custom UERANSIM docker image. This custom image includes scripts from Puppeteer, which enables the UERANSIM test-suite to simulate a UE performing normal user activities (e.g. watching YouTube, browsing websites, etc.) while connected to a 5g network.

• In the event that the custom image(s) is not stored in an existing image repository, please create a new AWS ECR repository for it to be uploaded & stored:
  • Step 1 - Create an ECR repository
  • Step 2 - In the ".gitlab-ci.yml" file, change line 5 to represent the repository registry URI for your AWS ECR repository

Prerequisites:

• Please ensure that you have configured the AWS CLI to authenticate to an AWS environment where you have adequate permissions to create an EKS cluster, security groups and IAM roles
• Please ensure that there is an existing image repository where these custom docker images can be stored.

Steps:

1. Create a Gitlab account

2. Import this repository into Gitlab

3. Perform a "Git clone" of this repository on your local machine

4. Set up a private Gitlab runner on the CNTF EKS cluster (Note: You only need to do this process once, this runner can be used by the other CNTF repositories you execute):

   • In Gitlab, on the left side of the screen, hover over "settings" and select "CI/CD"
   • Next to "Runners" select "expand"
   • Unselect "Enable shared runners for this project"
   • Click "New project runner"
   • Under "Operating systems" select "Linux"
   • Fill out the "Tags" section and select "Run untagged jobs"
   • Scroll to the bottom and select "Create runner"
   • Copy and save the "runner token" listed under "Step 1"
   • Select "Go to runners page", you should now see your runner listed with a warning sign next to it under "Assigned project runners"
   • On your local terminal:
     • Install the helm gitlab repository: helm repo add gitlab https://charts.gitlab.io
     • intialize helm (for helm version 2): helm init
     • create a namespace for your gitlab runner(s) in the cntf cluster: kubectl create namespace <NAMESPACE (e.g. "gitlab-runners")>
     • Install your created runner via helm: helm upgrade --install <RUNNER_NAME> -n <NAMESPACE> --set runnerRegistrationToken=<RUNNER_TOKEN> gitlabUrl=http://www.gitlab.com gitlab/gitlab-runner
     • Check to see if your runner is working: kubectl get pods -n <NAMESPACE> (you should see "1/1" under "READY" and "Running" under "STATUS")
     • Give your runner cluster-wide permissions: kubectl apply -f gitlab-runner-rbac.yaml
   • In Gitlab, Under "Assigned project runners" you should now see that your runner has a green circle next to it, signaling a "ready" status
   • How to re-use this runner for other CNTF repositories:
     • Hover over "Settings" and select "CI/CD"
     • Under "Other available runners", find the runner you have created and select "Enable for this project"

5. Authenticate Gitlab with AWS:

   • Inside of the Gitlab repository, hover over "Settings" and select "CI/CD"
   • Next to "Variables" select "Expand"
   • Select "Add variable"
   • Under "Key" type: AWS_ACCESS_KEY_ID
   • Under "Value", enter the value of your AWS Access key
   • Repeat steps 3-5 for AWS_SECRET_ACCESS_KEY and AWS_SESSION_TOKEN (optional) with the correct corresponding values

6. Run the CI/CD pipeline:

   • On the left side of the screen click the drop-down arrow next to "Build" and select "Pipelines"
   • In the top right hand corner select "Run Pipeline"
   • In the drop-down under "Run for branch name or tag" select the appropriate branch name and click "Run Pipeline"
   • Once again, click the drop-down arrow next to "Build" and select "Pipelines", you should now see the pipeline being executed

Goal of each stage in the pipeline (refer to ".gitlab-ci.yml" for more details):

• "build" - builds the customer UERANSIM docker image and pushes it to an AWS ECR repository

└── README.md
|
|
└── .gitlab-ci.yml             contains configurations to run CI/CD pipeline
|
|
└── Dockerfile                 contains the configurations to create the custom ueransim docker image
|   
|
└── amazon-search.js           contains configurations to simulate a search on amazon.com and can be inserted into the dockerfile to modify the ueransim test-suite
|
|
└── devtools-search.js         contains configurations to simulate a search on dev-tools.ai and can be inserted into the dockerfile to modify the ueransim test-suite
|
|
└── youtube-search.js          contains configurations to simulate watching a video on youtube.com and can be inserted into the dockerfile to modify the ueransim test-suite