Overview

Run an SSH script remotely. In particular we will:

  • Create a program on Pristine Lubuntu
  • From that program we will run the CpuInfo script on our EC2 instance and display the output in our client.
    • Recall that we copied that file to our Ec2 instance in the assignment called EC2 Copy File.

Get Started

Make sure you have pulled the most recent version of JsObjects.

jo
git pull

Remember that you can get back to the directory you were in before typing jo by typing cd -.

In the root of your repository, create a directory called week06-run-ssh and navigate into it.

From the root of your new directory, run the get-gist program by typing its name at the command prompt and hitting the enter key. Select Elven Create Concurrently which is probably option l.

Run the elf-concur script. When prompted, choose the server option on 30026 from the menu. It should create the client and server directories, configure them, and set up concurrently in the week06-run-ssh directory.

The elf-concur script should have set your programs up so that HTTP commands issued by the client will be proxied to the server. It did this by placing a proxy property in package.json. In short, the programs were set up for you automatically.

Navigate to the client directory. Run get-gist a second time. Select:

a) Run ESLintRc and Prettier (cdef)

Exit get-gist. Run the program again in the server directory, again selecting the ESLintRc option from the get-gist menu.

From the WebStorm menu, choose **File Settings Languages & Frameworks JavaScript Code Quality Tools**. Enable ESLint and disable the other tools such as JSCS and JsHint.

Navigate back to the root of the week05-run-ssh directory. Run npm start. The client and server should now be running on the appropriate ports.

  • See /server/bin/www for the set up of the port for the server.
  • See /client/package.json for the proxy, which is at the bottom of the file.
  • See week05-run-ssh for the set up of concurrently
    • package.json
    • start-client.js

Get Live Templates

Open the project in WebStorm.

At the bash shell, issue this command to download the latest Elven WebStorm Live Templates:

wget https://s3.amazonaws.com/bucket01.elvenware.com/images/elf-webstorm-settings.jar
Choose **File Import Settings** from the WebStorm menu. Select the jar file you just downloaded.

NOTE: If you start creating your own Live Templates, I strongly suggest that you export them frequently so that you do not lose any of your work. I don’t think importing live templates deletes existing ones, but I am not sure.

Snippets in Atom and Code

In Atom, choose **Edit Snippets**

Here is an HTML example:

'.source.html':
  'elf-table':
    'prefix': 'Elf table'
    'body': 'table>(thead>tr>(th*3))*(tbody>tr>(td*3))'
In VS Code, I went to **File Preferences User Snippets** and found html.json. Like this:
"Create elf table": {
		"prefix": "elf-table",
		"body": ["table>(thead>tr>(th*3))*(tbody>tr>(td*3))"],
		"description": "Create elf table"
	}

For VS Code, where they want one line at a time, RegEx to put each line in quotes with comma

  • Find: (^\s)(.)
  • Replace: \1"\2",

Header in render Method

We can view the templates by selecting **File Settings Editor Live Templates user. Note that some of the templates only work in certain places. For instance, the JSX related templates will only run if your cursor is in the appropriate part of, for instance, a react **render method.

Let’s test our new Live Template. Open up src/App.js. In the render method, remove the JSX HTML header. Type elf-jsx-start to set up one of our live templates. Press tab or enter to invoke it. The following code should be inserted automatically into your code:

<header>
    <h1></h1>
    <p className="byline">by </p>
</header>
<main>

</main>
<footer>
    <p>&copy; 2018 </p>
</footer>

Fill in the missing pieces. For instance, insert your name in the appropriate places, and add some text for the H1 element. Suggested text might Run SSH.

Button

In the main element run the elf-react-button live-template:

<button onClick={this.}></button>

Fill in the missing pieces.

Call Fetch

Above the render method but inside the App class declaration, run the elf-call-fetch live-template:

= () => {
   fetch('/')
       .then(function (response) {
           return response.json();
       })
       .then(function (json) {
           console.log('JSON from server:', json);

       })
       .catch(function (ex) {
           console.log('parsing failed, error on server, URL bad, network down, or similar');
           console.log(JSON.stringify(ex, null, 4));
       });
};

Fill the missing pieces. The method should be called callCpuInfo. The endpoint URL for our route should be:

  • call-cpu-info

Server Side

Navigate to the server directory. We will need the ssh2 NPM package:

npm install ssh2 --save

At the top of routes/index.js require the package:

const Client = require('ssh2').Client;

Near the top of routes index.js put these declarations and method:


const hostAddress = '<YOUR ELASTIC_IP_HERE>';

let allData = '';

const runCpuInfo = (hostAddress, response) => {
    var conn = new Client();
    conn.on('ready', function() {
        console.log('Client :: ready');
        conn.exec('~/CpuInfo', function(err, stream) {
            if (err) throw err;
            stream
                .on('close', function(code, signal) {
                    console.log(
                        'Stream :: close :: code: ' +
                            code +
                            ', signal: ' +
                            signal
                    );
                    conn.end();
                    response.send({ result: 'success', allData: allData });
                })
                .on('data', function(data) {
                    console.log('STDOUT: ' + data);
                    allData += data;
                })
                .stderr.on('data', function(data) {
                    console.log('STDERR: ' + data);
                    allData += data;
                });
        });
    }).connect({
        host: hostAddress,
        port: 22,
        username: 'ubuntu',
        privateKey: require('fs').readFileSync(
            process.env.HOME + '/.ssh/<YOUR_PRIVATE_KEY_HERE>'
        )
    });
};

Put your elastic IP in the hostAddress variable. Put the name of your private key in the appropriate place near the bottom of the callCpuInfo method.

In routes/index.js also put an endpoint for call-cpu-info. The elf-router-get live-template should help you handle this. from that endpoint, call the method we declared above like this:

runCpuInfo(hostAddress, response);

Test your code and make sure it works. In the console, you should see the output from running the CpuInfo script on your EC2 server.

Create sshRunner Module

Create a file called routes/ssh-runner.js. Move the appropriate code for handling the call-cpu-info endpoint from routes/index.js into that file. Remove the redundant code from routes/index.js.

NOTE: In places, I may also refer to ssh-runner.js as run-ssh.js. They are both the same, or very similar files. Sorry about the confusion.

Turn it in

Specify:

  • branch (week06)
  • folder (week06-run-ssh)
  • tag (week06-run-ssh)
  • Repo URL

Include a screenshot of the result you see in your browser console when your run the script. If you can, show the bash terminal in the background.