remote-execution

The purpose of this tool is to help with the following three use cases:

1. You want to run a piece of code in a remote environment, but the code is too large or it is otherwise impractical to paste it into the console (sometimes pasting many lines of code breaks the session).

2. You want to pull information from a remote environment's console, but this information is too large to easily copy it from the console.

3. You want to do both 1 and 2 at the same time.

⚠️⚠️⚠️ WARNING ⚠️⚠️⚠️

This is not meant to be used in a production environment!

Dependencies

You only need Python 3.6+ and ngrok. As a reminder, ngrok can be installed via brew install ngrok, and you need to configure it. Here's a sample configuration for ngrok. It only takes one file:

# In mac: ~/Library/Application Support/ngrok/ngrok.yml
authtoken: "*****" # from https://dashboard.ngrok.com/get-started/your-authtoken
version: "2"
region: us

Use case 1

1. Put the code that you want to run in a file. Name the file something that starts with code (e.g. code.rb, code_pull_assignments.rb), and place it in this directory. For example:

   # code.rb
   class RemoteExecution
     def self.run
       # ... logic to run
     end
   end

2. Open a terminal session and start the remote-execution server with:

   python server.py

3. Open another terminal session and run ngrok:

   ngrok http 8000

   Take note of the https url as it is going to be used in the next step. This url looks something like https://c7fd6475f29d.ngrok.io. You could also specify the --hostname flag if you have one here, but beaware that this could pose a security risk if others know this url, as they can easily see all files in this directory.

   You need to specify 8000 because server.py runs on that port by default.

4. Open another terminal session to console into the desired environment, and eval the code from step 1 by pulling it via the ngrok url from step 2 and using the file name for the path. For example, if you're using ruby:

   eval(open('https://c7fd6475f29d.ngrok.io/code.rb').read); RemoteExecution.run
   eval(open('https://c7fd6475f29d.ngrok.io/code_pull_assignments.rb').read); RemoteExecution.run

That's it. If you want to test this locally, you can skip the ngrok step and just use http://127.0.0.1:8000 in step 4.

Use case 2

1. Open a terminal session and start the remote-execution server. See step from Use case 1 for details.

2. Open another terminal session and run ngrok. See step from Use case 1 for details.

3. Open another terminal session to console into the desired environment. Perform an HTTP POST request to the url from step 2 with the json-formatted information that you want to pull as the body of the request. For example, if you're on ruby:

   payload = {'data': 5} # information that you would like to pull
   HTTParty.post('https://c7fd6475f29d.ngrok.io', :body => payload.to_json)

   You can also test this directly from curl:

   curl -X POST https://c7fd6475f29d.ngrok.io -d '{"data": 6}'

Use case 3

1. Put the code that you want to run in a file. Name the file something that starts with code (e.g. code.rb, code_pull_assignments.rb), and place it in this directory. Include a request at the end of the code to send the information back. If you're running ruby code, this would look something like this:

   # code.rb
   class RemoteExecution
     def self.run
       payload = {} # information that you would like to pull
   
       # ... run code to populate `payload`
   
       HTTParty.post('https://c7fd6475f29d.ngrok.io', :body => payload.to_json)
     end
   end

2. Open a terminal session and start the remote-execution server. See step from Use case 1 for details.

3. Open another terminal session and run ngrok. See step from Use case 1 for details.

4. Open another terminal session to console into the desired environment, and eval the code from step 1 by pulling it via the ngrok url from step 2 and using the file name for the path. For example, if you're using ruby:

   eval(open('https://c7fd6475f29d.ngrok.io/code.rb').read); RemoteExecution.run