The Future Studio team uses Strider to automatically test and deploy the code for our homepage and blog theme. Even though we rely on Ghost for our blog, we keep our theme in a git repository and trigger the deployment right away after every push.
Previously published articles in this series explain how to connect with git hosting services and trigger automatic tests in Strider.
— TODO outline —
Strider offers the strider-ssh-deploy plugin for continuous deployment via SSH to any server. Install the plugin via Strider’s admin panel if it isn’t already installed.
Search for SSH Deploy and and make sure the plugin is installed. If not, click the Install button.
Head over to the settings page for the project you want to deploy via SSH. Click the Plugins option in the left sidebar. Now drag and drop SSH Deploy from Available Plugins into Active Plugins.
Pushing the SSH Deploy card into available plugins adds the plugin settings in Strider’s left sidebar. Click the newly available SSH Deploy settings link to define necessary deployment settings.
First, type the username of the destination host user. This is the user we use to log in via SSH and execute the final deployment. The example above uses the boring
ssh-deploy-user, please use a more precise username on your host.
Second, specifiy the destination host url with SSH port. The port is optionally if the SSH server on your destination host uses the default port 22. The SSH server on our examplary
yourhost.url runs on port
Third, define the commands to be executed on destination host. You can check the Transfer bundle option to move a
.tar.gz to the remote host. This bundle contains the repositories code as a compressed tarball.
The screen above illustrates code to execute on the remote system. We
cd into the transfered and uncompressed bundle folder,
pwd the current working directory, install dependencies with
npm install and start the node server with
pm2 node process monitor.
In order to authorize the provided SSH user on destination host, we need to make our project’s public key available in the
Visit the projects settings. This will show you private and public SSH keys for the repository. Strider will use this keys for authorization to other hosts and services.
Copy the public key.
Now use your shell and SSH into your destination host, switch the the user to the specified one in Strider for SSH deployment. From here, change the directory to the system users SSH folder (usually
Edit or create (if not existing) the
authorized_keys file in the SSH folder. Paste the copied Strider SSH public key into the
authorized_keys file. This allows Strider to connect to your destination host with the help of the corresponding private key.
The configuration is done. Now we can test the deployment pipeline :)
If you currently don’t have any code to push into the repository and trigger an automated build process, we can start the integration and deployment manuall from Strider’s dashboard.
Strider initiates the test process and performs the known integration. The difference to previous integrations: the deploy step performs your SSH deployment.
If everything went smooth, the deployment is available right after tests and deployment finish sucessfully. Visit the url of your server and check your deployed app.
This guide walks you through the process of SSH deployments with Strider to any server. With the previously published articles, you’re able to connect Strider with your git repositories, test the code and start deployments right after successful integration. Your pipeline assumes shape, Booyah!
Within the next blog post we show you how to integrate Slack and notify channels about build and deployment statuses.
If you need help, don’t hesitate to use the comments or contact us via @futurestud_io.