CI/CD

Version Control.

Dash apps are written in Python code so you can use version control and platforms like GitHub/GitLab to manage releases, review each other’s work, write tests, and rollback to old versions. This level of version control and rollback is not possible with point-and-click BI analytics.

Dash App Continuous Deployment.

Deploying a change to a Dash App is dead-simple and entirely programatic. By running the command git push plotly master, your version-controlled modifications are pushed to Dash Enterprise and the application is automatically rebuilt with zero-downtime. With programatic deployments and SSH key authentication, you can easily integrate the git push deployment command into your CI/CD pipelines. Learn more about how Dash Enterprise builds and deploys Dash Apps: https://dash.plotly.com/dash-enterprise/application-structure.

.

Review Apps.

Review Apps are “pre-release” versions of your Dash Apps that are created automatically by your CI system in a pull request. Review Apps allow Dash App developers to instantly share the changes that they propose to their team members before those changes go live. Instead of speculating on how a change to your Dash App might run, you can actually see the changes running in a live environment. Review Apps are created at a live, temporary, shareable URL on Dash Enterprise and are automatically deleted after the proposed change goes live. Review Apps automatically inherit all of the settings of the app in production so that are certain certain that you’re testing against the same situations your code will face after you merge into production.

GitOps Server Admin.

IT and DevOps admins that manage and configure Dash Enterprise can configure the platform with a GitOps workflow. Instead of configuring the application using the built-in Server Manager user interface, you can manage the settings entirely through a git-based CI/CD workflow. This includes platform configuration changes like the Authentication Middleware settings, license updates, and other settings. (Note that the Dash apps created by data scientists are already managed via git.)

Automated Testing.

The Dash App framework includes a built-in testing capabilities for browser-based integration tests. Use dash.testing to test the UI of your application in your existing CI/CD pipeline.

With zero-downtime deployments, run custom health checks during the deployment process to prevent critical bugs from making their way into production.