f7c129340a | ||
---|---|---|
.. | ||
api | ||
api-test | ||
app | ||
meshcentral | ||
nginx-proxy | ||
salt | ||
.env.example | ||
docker-compose.dev.yml | ||
docker-compose.override.yml | ||
docker-compose.test.yml | ||
docker-compose.yml | ||
readme.md |
readme.md
Docker Setup
- install docker and docker-compose
- Obtain wildcard cert or individual certs for each subdomain
Generate certificates with certbot
Install Certbot
sudo add-apt-repository ppa:certbot/certbot
sudo apt-get install certbot
Generate the wildcard certificate. Add the DNS entry for domain validation.
sudo certbot certonly --manual -d *.example.com --agree-tos --no-bootstrap --manual-public-ip-logging-ok --preferred-challenges dns
Copy the fullchain.pem and privkey.pem to the nginx-proxy/cert directory.
Configure DNS and Firewall
You will need to add DNS entries so that the three subdomains resolve to the IP of the docker host. There is a reverse proxy running that will route the hostnames to the correct container. On the host, you will need to ensure the firewall is open on tcp ports 80, 443, 4505, 4506.
Run the environment with Docker
Copy the .env.example to .env then change values in .env to match your environment
cd docker
sudo docker-compose up -d
You may need to run this twice since some of the dependant containers won't be ready
Create a super user
sudo docker-compose exec api python manage.py createsuperuser
Setup 2FA authentication
Get the 2FA code with
sudo docker-compose exec api python manage.py generate_totp
Use the generated code and the username to generate a bar code for your authenticator app (domain is the domain name of your site, for example: rmm.example.com)
sudo docker-compose exec api python manage.py generate_barcode [2FAcode] [username] [domain]
Rebuild the api container
sudo docker-compose up -d --build api
Get MeshCentral EXE download link
Run the below command to get the download link for the mesh central exe. The dashboard will ask for this when you first sign in
sudo docker-compose exec api python manage.py get_mesh_exe_url
Connect to a container instance shell
The below command opens up a shell to the api service.
sudo docker-compose exec api /bin/bash
If /bin/bash doesn't work then /bin/sh might need to be used.
Using Docker for Dev
This allows you to edit the files locally and those changes will be presented to the containers. Hot Module Reload (Vue/webpack) and the Python equivalent will also work!
Setup
Files that need to be manually created are:
- api/tacticalrmm/tacticalrmm/local_settings.py
- web/.env
Make sure to add MESH_WS_URL="ws://meshcentral:443"
in the local_settings.py file. This is needed for the mesh central setup
For HMR to work with vue you can copy .env.example and modify the setting to fit your dev environment.
Create Python Virtual Env
Each python container shares the same virtual env to make spinning up faster. It is located in api/tacticalrmm/env.
There is a container dedicated to creating and keeping this up to date. Prior to spinning up the environment you can run docker-compose -f docker-compose.dev.yml up venv
to make sure the virtual env is ready. Otherwise the api and celery containers will fail to start.
Spinup the environment
Now run docker-compose -f docker-compose.yml -f docker-compose.dev.yml up -d
to spin everything else up
This will mount the local vue and python files in the app container with hot reload. Does not require rebuilding when changes to code are made and the changes will take effect immediately!
Running the Tests
There is a container that is dedicated to run the vue unit tests. The below command will run them and display the output. You can ignore the orphaned containers message.
docker-compose -f docker-compose.test.yml up app-unit-test
Other Considerations
-
Using Docker Desktop on Windows will provide more visibility into which containers are running. You also can easily view the logs for each container in real-time, and view container environment variables.
-
If you are on a *nix system, you can get equivalent logging by using
docker-compose logs [service_name]
. -
docker ps
will show running containers. -
docker system prune
will remove items that are not in use by running containers. There are also--all and --volumes
options to remove everything if you want to start over. Stop running containers first.docker-compose -f docker-compose.yml -f docker-compose.dev.yml down
-
If the docker container isn't getting file changes you can restart the host or do a
docker system prune --volumes
. This will remove the docker volumes and will create a new one once the containers are started.