We're going to move to docker as the recommended launch method. Have a docker install script to customize the docker compose file, make sure the user is in the dial out group, and detect the radio Port by ID (or accept TCP, and warn that bluetooth is not recommended for docker, but allow the user to proceed). Do all the similar stuff we do in the persistent service setup script -- ask for transport info, or autodetect (probably need to actually look at serial devices and confirm by ID, or ask for the path). Figure out if websockets over docker needs something special. Add some environment variable (?) or easy key generation for snakeoil https cert that gets passed in (probably need to .gitignore those if we're generating them somewhere visible). Supersedes comments in #119.
We're going to move to docker as the recommended launch method. Have a docker install script to customize the docker compose file, make sure the user is in the dial out group, and detect the radio Port by ID (or accept TCP, and warn that bluetooth is not recommended for docker, but allow the user to proceed). Do all the similar stuff we do in the persistent service setup script -- ask for transport info, or autodetect (probably need to actually look at serial devices and confirm by ID, or ask for the path). Figure out if websockets over docker needs something special. Add some environment variable (?) or easy key generation for snakeoil https cert that gets passed in (probably need to .gitignore those if we're generating them somewhere visible). Supersedes comments in #119.
<!-- github-sync-id:4166612978 -->
Move existing docker compose to docker-compose.example.yml and have copying be part of the setup. gitignore docker-compose.yml so we don't steamroll people who have already customized their docker-compose.yml, then make the new docker launch script generate a good version.
Move existing docker compose to docker-compose.example.yml and have copying be part of the setup. gitignore docker-compose.yml so we don't steamroll people who have already customized their docker-compose.yml, then make the new docker launch script generate a good version.
<!-- github-comment-id:4158574385 -->
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
We're going to move to docker as the recommended launch method. Have a docker install script to customize the docker compose file, make sure the user is in the dial out group, and detect the radio Port by ID (or accept TCP, and warn that bluetooth is not recommended for docker, but allow the user to proceed). Do all the similar stuff we do in the persistent service setup script -- ask for transport info, or autodetect (probably need to actually look at serial devices and confirm by ID, or ask for the path). Figure out if websockets over docker needs something special. Add some environment variable (?) or easy key generation for snakeoil https cert that gets passed in (probably need to .gitignore those if we're generating them somewhere visible). Supersedes comments in #119.
Move existing docker compose to docker-compose.example.yml and have copying be part of the setup. gitignore docker-compose.yml so we don't steamroll people who have already customized their docker-compose.yml, then make the new docker launch script generate a good version.
Do we use the hostname anywhere internally? Will that cause any problems inside docker?