Start by logging in to Palette. From the landing page, click on the user drop-down Menu and click on App Mode.

From the app mode landing page, navigate to the left Main Menu and click on Virtual Clusters. Next, click on the button New Virtual Cluster.

In the following screen, you will be prompted for the cluster group, virtual cluster name, and the cluster size in terms of CPU, memory, and storage. Select beehive for the cluster group, name the cluster cluster-1, and allocate 4 CPU, 4 GiB memory, and 2 GiB of storage. Click on Deploy Virtual Cluster after you have filled out all the required information.

Palette’s Dev Engine allows you to deploy up to two virtual clusters into the Beehive cluster group. Each virtual cluster requires a minimum of 4 CPU, 4 GiB memory, and 2 GiB storage. When using the Beehive cluster, you can allocate a maximum of 12 CPU, 16 Gib memory, and 20 GiB of storage. Check out the Palette Dev Engine and Quotas documentation to learn more about limits.

It will take a few minutes for the virtual cluster to deploy. In the meantime, navigate to the left Main Menu and click on App Profiles.

App Profiles are templates that contain all the configurations and settings required to deploy applications to virtual clusters. App Profiles provide a way to drive consistency across virtual clusters as you can re-use app profiles and deploy them to different virtual clusters. You can think of app profiles as declarative templates that inform the Kubernetes cluster of the desired application or set of applications.

Click on the New App Profile button to start creating your first app profile. Give the app profile the name hello-universe-ui and add the tag scenario-1. Click on Next. The following screen is the service type selection page. You have the option to deploy applications through containers, Helm, or Manifests. You can also consume services such as databases and more. Click on Container Deployment.

Name the container ui, select a public registry, and provide the image URL ghcr.io/spectrocloud/hello-universe:1.0.10. Change the network access to Public and add the port 8080.

Click on Review once you have filled out the provided information. On the next page, click on the Deploy New App button.

It’s time to deploy your application to a virtual cluster. Name the application single-scenario. For the App profile input field, click on the button to the right of the input field to get a list of all your available app profiles. Select the hello-universe-ui profile and click on Confirm.

Next, click the radio button Deploy in An Existing Palette Virtual Cluster. Select cluster-1 and click on Create App to deploy the app profile onto the virtual cluster.

The app profile deployment takes a few moments to finish. You can review the application's deployment progress by navigating to the left Main Menu and selecting Virtual Clusters. Click on cluster-1 to view its details page. You can review cluster information, log events, access a remote shell session in the cluster, and more from the cluster details page.

When the application is deployed and ready for use, the Services row on the details page will automatically be updated by Palette with the app's public-facing URL. Click on the :8080 link to view the application.

Welcome to Hello Universe, a demo application to help you learn more about Palette and its features. Feel free to click on the logo to increase the global counter and for a fun image change.

You have deployed your first application to Palette. Your first application is a single container application with no upstream dependencies. In a production environment, you often deploy applications that consume other services and require connectivity with other resources. The next scenario expands on the single application scenario by adding an API server and Postgres database to simulate a common application architecture encountered in a production environment.

Create another virtual cluster for the multi-application scenario. From the app mode landing page, navigate to the left Main Menu and click on Virtual Clusters. Next, click on the New Virtual Cluster button.

Add the following details. Select beehive for the cluster group, name the cluster cluster-2, add the tag scenario-2, and allocate 8 CPU, 12 GiB memory, and 12 GiB of storage. Click on Deploy Virtual Cluster after you have filled out all the required information.

It will take a few minutes for the new virtual cluster to deploy. In the meantime, go ahead and navigate to the left Main Menu and click on App Profiles.

Click on the New App Profile button to create your second app profile. Give the app profile the name hello-universe-complete and add the tag scenario-2. Click on Next. This application profile will contain three different applications, and you will create a service configuration for each. The three layers or tiers will together make up the entire application deployment. The order in which you create each layer plays an important role, as it dictates the deployment order. For this scenario, you will deploy the database, the API, and the UI. To create the first layer, select the database service Postgres.

In the next screen, assign the following values to the Postgres database.

• Name: postgres-db

• Username: pguser

• Database Name: counter

• Database Volume Size: 2

• Version: 14

Take note of the Output Variables section. The Postgres service exposes several output variables to help other applications connect with the database. In the next section, you will use these output variables and other output variables that Palette exposes for each service. You can learn more about output variables by reviewing the app profile output variables documentation.

Next, navigate to the top left side of the wizard screen and click on the Actions button +. Go ahead and select Container Deployment.

The API is available as a container image. To deploy the API successfully, you need to provide the API server with information about the database such as hostname, database user, database name, and password. The required information can be retrieved using Palette's global output variables and the output variables the database service exposes.

Provide the container service with the following information:

• Container Name: api

• Registry: Public

• Image: ghcr.io/spectrocloud/hello-universe-api:1.0.8

• Network Access: Public

• Ports: 3000

Assign the following environment variables to the API service:

You can learn more about each environment variable's purpose by reviewing the API server's documentation. One variable that you should understand in greater detail is the DB_HOST. The value of this environment variable is constructed using the output variables the Postgres service exposed. The {{.spectro.app.$appDeploymentName.postgres-db.POSTGRESMSTR_SVC}} variable contains the Kubernetes DNS value of the Postgres service container.

A virtual cluster is a Kubernetes environment, and because it’s a Kubernetes environment, you can use the Kubernetes DNS record created for each service and pod. You will have another opportunity to practice this concept when you deploy the UI.

When you have filled out all the required information, navigate to the top left side of the wizard screen and click on the Actions button +. Select the Container Deployment to add the final service layer, the UI.

This time the UI will point to the API server that you manage. The API server has authentication enabled, so to ensure all API requests are accepted you will provide the UI with the anonymous token.

Provide the UI container with the following information.

• Container Name: ui

• Registry: Public

• Image: ghcr.io/spectrocloud/hello-universe:1.0.10

• Network Access: Public

• Ports: 8080

Assign the following environment variables to the UI service:

If you want to explore the UI service's environment variables in greater detail, you can review the UI documentation. The API_URI contains the address of the application load balancer that will be deployed for the API service.

The output variable {{.spectro.app.$appDeploymentName.api.CONTAINER_SVC_EXTERNALHOSTNAME}} is used to retrieve the load balancer URL value.

Click on the Review button at the bottom of the screen to finalize the app profile. Click on Deploy New App in the following screen to deploy the new app profile to cluster-2.

Name the app multiple-app-scenario, select the app profile hello-universe-complete, pick version 1.0.0 and toggle the radio button Deploy In An Existing Palette Virtual Cluster. Select cluster-2 and click on Create App.

The app profile deployment takes a few moments to finish. You can review the application's deployment progress by navigating to the left Main Menu and selecting Virtual Clusters. Click on cluster-2 to view its details page.

Once the app is successfully deployed, the cluster details page will expose the public-facing URLs of the services.

Click on the UI’s service URL for port 8080 to access the Hello Universe application in a three-tier configuration.

The global counter is no longer available. Instead, you have a counter that starts at zero. Each time you click on the center image, the counter is incremented and stored in the Postgres database along with metadata. Also, remember that the reverse proxy injects the Bearer token value in each request sent to the API.

To remove all resources created in this tutorial, begin by navigating to the left Main Menu and select Apps. For each application, click on the three-dots Menu to expand the options menu and click on the Delete button. Repeat this process for each application.

Next, in the left Main Menu, click on the Cluster link to access the clusters page.

Click on cluster-1 to access its details page. Click on Settings from the details page to expand the settings menu. Click on Delete to delete the cluster. You will be asked to enter the cluster name to confirm the delete action. Type the cluster name to proceed with the delete step. Repeat this process for cluster-2.