TNLCM has been designed as a modular application, with the intention of making certain components easily replaceable or extended, while minimizing the effect of changes in other parts of the application. At the same time, there is an emphasis on re-usability, where several data structures and generic logic can be shared between the different components of the application.
Table of Contents
- Programming language.
- Python framework for web applications to expose the API.
- NoSQL database designed to store Trial Networks.
TNLCM/ // main folder.
├─ .github/ // folder contains files and templates for GitHub workflow automation.
├─ conf/ // folder that handler the configuration files.
├─ core/ // folder that the developed code is stored.
│ ├─ auth/ // folder that handler the authentication of users who have access.
│ ├─ cli/ // folder that handler the cli for run commands.
│ ├─ database/ // folder that handler the connection with MongoDB database using mongoengine.
│ ├─ exceptions/ // folder that handler the creation of custom exceptions.
│ ├─ jenkins/ // folder that handler the connection with Jenkins for tn deployment.
│ ├─ logs/ // folder that handler the logs configuration.
│ ├─ mail/ // folder that handler the configuration to use flask mail library.
│ ├─ models/ // folder that contains the database models.
│ ├─ repository/ // folder that handler the connection to any repository.
│ ├─ routes/ // folder that handler the API that is exposed.
│ ├─ sixg_library/ // folder that handler the connection to the 6G-Library repository.
│ ├─ sixg_sandbox_sites/ // folder that handler the connection to the 6G-Sandbox-Sites repository.
│ ├─ temp/ // folder that handler the creation of temporary files.
│ └─ utils/ // folder that handler data conversions and storage in YAML, Markdown, and JSON formats.
├─ docker/ // folder containing Docker-related configuration and setup files.
├─ docs/ // folder where all documentation is stored.
├─ scripts/ // folder contains scripts for automated deployments.
│ ├─ deploy_vm.sh // file to deploy TNLCM with its dependencies on a vm.
│ └─ deploy_docker.sh // file to deploy TNLCM in docker with its dependencies.
├─ tn_template_lib/ // folder that trial network descriptors templates are stored.
├─ .dockerignore // file specifying which files and directories to ignore in the Docker build context.
├─ .env.template // file that contains placeholder environment variables for configuring the application.
├─ .gitignore // file specifying intentionally untracked files to ignore.
├─ app.py // main file that starts TNLCM.
├─ CHANGELOG.md // file containing the changes made in each release.
├─ docker-compose.yaml // file for create database.
└─ requirements.txt // file containing the libraries and their versions.
TNLCM is a state machine that allows the automation of component deployment. Green indicates what is implemented and red indicates what is in the process of implementation.
- Validated: trial network descriptor created is checked to see if it can be deployed. 6G-Library is used to validate it
- Activated: trial network has been deployed in OpenNebula
- Suspended: trial network has been suspended. It remains deployed, but turned off
- Failed: there was an error during the trial network deployment
- Destroyed: trial network deployment in OpenNebula is removed, but its content is kept in the database and locally
- Purge: the trial network and the callbacks are removed from the database as well as its local content
- Initial state → Validated: trial network descriptor validated and ready for deploy
- Validated → Activated: trial network deployed and ready for use
- Validated → Failed: trial network deployment failed
- Validated → Purge: trial network invalid
- Failed → Failed: again, trial network deployment failed
- Activated → Destroyed: trial network destroyed and ready for deploy again
- Activated → Suspended: TODO
- Suspended → Activated: TODO
- Destroyed → Activated: trial network deployed and ready for use
- Destroyed → Purge: trial network removed
Note
TNLCM is being developed and tested on Ubuntu in version 24.04 LTS.
Important
TNLCM requires the prior installation of:
| Repository | Release |
|---|---|
| OpenNebula | v6.10 |
| Jenkins | v2.462.3 |
| MinIO | 2024-07-04 |
Tip
Additionally, TNLCM depends on:
| Repository | Branch | Release |
|---|---|---|
| 6G-Library | - | v0.3.0 |
| 6G-Sandbox-Sites | platform | - |
Download the installation script which is deploy_vm.sh and is located in the scripts folder.
Update the script and add the contents of the following variables:
Important
There is a comment TODO in the script.
TNLCM_ADMIN_USERTNLCM_ADMIN_PASSWORDTNLCM_HOSTJENKINS_HOSTJENKINS_USERNAMEJENKINS_PASSWORDJENKINS_TOKENSITES_TOKEN
Note
Execute the script with the root user.
Once the environment variables have been filled in, run the script:
chmod 777 deploy_vm.sh./deploy_vm.shA MongoDB dashboard will be available at the url http://mongo-express-ip:8081 where the database can be managed.
Note
User and password to access to the MongoDB dashboard are the values indicated in the variables ME_CONFIG_BASICAUTH_USERNAME and ME_CONFIG_BASICAUTH_PASSWORD of the .env file. By default, the values indicated in the .env.template file are used.
gunicorn -c conf/gunicorn_conf.pyA Swagger UI will be available at the url http://tnlcm-backend-ip:5000 where the API with the endpoints can be seen.
Warning
The format of Trial Network Descriptors has not been finalized and is expected to change in the future.
Trial Network Descriptors are yaml files with a set of expected fields and with the following structure:
trial_network: # Mandatory, contains the description of all entities in the Trial Network
type-name: # Mandatory, entity name. Unique identifier for each entity in the Trial Network
type: # Mandatory, 6G-Library component type
name: # Mandatory, custom name. Not use character \- or \. Exclude components tn_init, tn_bastion and tn_vxlan
debug: # Optional, param to debug component in Jenkins. Possible values true or false
dependencies: # Mandatory, list of dependencies of the component with other components
- type-name
- ...
input: # Mandatory, dictionary with the variables collected from the input part of the 6G-Library
key: valueThis repository contains a variety of descriptors. Access the documentation to see what is defined in each of them.
08_descriptor.yaml- should work on all platforms. First end-to-end trial network.referente_tn.yaml- should work on all platforms. Reference trial network defined in WP5.
The API set forth in the TNLCM is as follows:
If it is the first time using the API it is necessary to create a user. A verification code is required, so it is necessary to enter a valid email address that can be accessed:
Once the registration code is obtained, proceed to the user registration using the email and the code previously employed:
Once the user has been created or if it has been previously created, add the user and its password in the green Authorize box:
Once the user has been added, an access token and its refresh token can be generated. This access token has a duration of 45 minutes (can be modified):
The next step is to add the token in the green Authorize box. It is required to put the word Bearer, a space and then the token. An example is shown:
Now, requests that involve having an access token can be made.
To create a Trial Network, the following endpoint must be used:
Once created, it will return a tn_id that will be required to deploy it.
To deploy a trial network, the following endpoint must be used:
If the access token expires, it can be refreshed by using the refresh token. The token in the green Authorize box must be updated with the refresh token and the post request must be made:
When the request is made, it will return another access token that will need to be put back into the green Authorize box.
The TNLCM database consists of several collections that store important information about trial networks, users, and verification tokens. Below is the description of each collection:
| Field | Description |
|---|---|
tn_id |
The ID of the trial network. |
entity_name |
The component_type-custom_name of component deployed. |
component_type |
Type of component deployed. |
custom_name |
Significant name of what is deployed. |
success |
Verification of successful or unsuccessful pipeline operation. |
markdown |
The entity_name report. |
output |
The data received by Jenkins decoded for the entitiy_name. |
| Field | Description |
|---|---|
component |
The component over which the resources are controlled. |
tn_id |
The ID of the trial network. |
quantity |
The amount of component available. |
ttl |
The amount of time the component can be used in a trial network. |
| Field | Description |
|---|---|
user_created |
The user that created the trial network. |
tn_id |
The ID of the trial network. |
state |
The state of the trial network. |
date_created_utc |
The date and time when the trial network was created (UTC). |
raw_descriptor |
The raw descriptor of the trial network. |
sorted_descriptor |
The sorted descriptor of the trial network. |
deployed_descriptor |
The current status of descriptor with the last entity deployed of the trial network. |
report |
The report related to the trial network. |
directory_path |
The directory where trial network is saved. |
jenkins_deploy_pipeline |
The pipeline used for the deployment of the descriptor. |
jenkins_destroy_pipeline |
The pipeline used for destroy a trial network. |
deployment_site |
The site where the trial network has been deployed. |
github_6g_library_commit_id |
The commit id of 6G-Library (branch, commit or tag) used to deploy trial network. |
github_6g_sandbox_sites_commit_id |
The commid id of 6G-Sandbox-Sites (branch, commit or tag) used to deploy trial network. |
| Field | Description |
|---|---|
email |
The email address of the user. |
username |
The username of the user. |
password |
The password of the user (hashed). |
org |
The organization to which the user belongs. |
| Field | Description |
|---|---|
new_account_email |
The email associated with the new account. |
verification_token |
The verification token generated for the new account. |
creation_date |
The creation date of the verification token. |
In the marketplace-community repository, which is a fork of OpenNebula, detailed information about the TNLCM appliance can be found.
To deploy 6G-SANDBOX TOOLKIT in OpenNebula, the documentation can be accessed from the web site.
| Photo | Name | GitHub | ||
|---|---|---|---|---|
 |
Carlos Andreo López | - |  |
 |
 |
Bruno García García | - | |
|