AUTOLab Machine Stats

This repo is forked from Hivemind for Berkeley EECS instructional computers.

How does it work?

Every five minutes, backend/census.py is executed. It connects to each server listed in backend/server.txt via SSH and collects information. The results from all of the servers are combined into a single JSON file (data/latest.json).

Overall load formula

The "overall load" heuristic is implemented in toRating() in main.js.

Contributing

If you would like to add / remove servers from the list, please file an issue or a pull request with your requested changes.

The current list of servers is at [backend/servers.txt][servers.txt].

Development

Want to host the website locally? Clone this repo, and start a web server in the project root directory.

The backend (i.e. the script that grabs data from the servers) is a little harder to set up:

0. Clone this repo.
1. Run make venv to create a virtualenv and install paramiko. You can also run pip install paramiko, but you'll probably want to install it locally to avoid polluting your system libraries, so a virtualenv works well for that.
2. Create an RSA key pair with no passphrase, rename the private key to hivemind_rsa and the public key to hivemind_rsa.pub and put them inside your home directory's SSH directory (~/.ssh).
3. Add the public key to your class account's ~/.ssh/authorized_keys file to allow hivemind to log in to the servers automatically.
4. Change the value of LOGIN_USERNAME in census.py to your login.

You should then be able to execute census.py to grab data from each server in servers.txt. The results are printed to stdout, which run_census puts into a file for the frontend to fetch.