Multi-App Run overview

Run multiple applications with one CLI command

Note

Multi-App Run for Kubernetes is currently a preview feature.

Let’s say you want to run several applications locally to test them together, similar to a production scenario. Multi-App Run allows you to start and stop a set of applications simultaneously, either:

  • Locally/self-hosted with processes, or
  • By building container images and deploying to a Kubernetes cluster
    • You can use a local Kubernetes cluster (KiND) or one deploy to a Cloud (AKS, EKS, and GKE).

The Multi-App Run template file describes how to start multiple applications as if you had run many separate CLI run commands. By default, this template file is called dapr.yaml.

Multi-App Run template file

When you execute dapr run -f ., it starts the multi-app template file (named dapr.yaml) present in the current directory to run all the applications.

You can name template file with preferred name other than the default. For example dapr run -f ./<your-preferred-file-name>.yaml.

The following example includes some of the template properties you can customize for your applications. In the example, you can simultaneously launch 2 applications with app IDs of processor and emit-metrics.

  1. version: 1
  2. apps:
  3. - appID: processor
  4. appDirPath: ../apps/processor/
  5. appPort: 9081
  6. daprHTTPPort: 3510
  7. command: ["go","run", "app.go"]
  8. - appID: emit-metrics
  9. appDirPath: ../apps/emit-metrics/
  10. daprHTTPPort: 3511
  11. env:
  12. DAPR_HOST_ADD: localhost
  13. command: ["go","run", "app.go"]

For a more in-depth example and explanation of the template properties, see Multi-app template.

Locations for resources and configuration files

You have options on where to place your applications’ resources and configuration files when using Multi-App Run.

Point to one file location (with convention)

You can set all of your applications resources and configurations at the ~/.dapr root. This is helpful when all applications share the same resources path, like when testing on a local machine.

Separate file locations for each application (with convention)

When using Multi-App Run, each application directory can have a .dapr folder, which contains a config.yaml file and a resources directory. Otherwise, if the .dapr directory is not present within the app directory, the default ~/.dapr/resources/ and ~/.dapr/config.yaml locations are used.

If you decide to add a .dapr directory in each application directory, with a /resources directory and config.yaml file, you can specify different resources paths for each application. This approach remains within convention by using the default ~/.dapr.

Point to separate locations (custom)

You can also name each app directory’s .dapr directory something other than .dapr, such as, webapp, or backend. This helps if you’d like to be explicit about resource or application directory paths.

Logs

The run template provides two log destination fields for each application and its associated daprd process:

  1. appLogDestination : This field configures the log destination for the application. The possible values are console, file and fileAndConsole. The default value is fileAndConsole where application logs are written to both console and to a file by default.

  2. daprdLogDestination : This field configures the log destination for the daprd process. The possible values are console, file and fileAndConsole. The default value is file where the daprd logs are written to a file by default.

Log file format

Logs for application and daprd are captured in separate files. These log files are created automatically under .dapr/logs directory under each application directory (appDirPath in the template). These log file names follow the pattern seen below:

  • <appID>_app_<timestamp>.log (file name format for app log)
  • <appID>_daprd_<timestamp>.log (file name format for daprd log)

Even if you’ve decided to rename your resources folder to something other than .dapr, the log files are written only to the .dapr/logs folder (created in the application directory).

Watch the demo

Watch this video for an overview on Multi-App Run:

Multi-App Run template file

When you execute dapr run -k -f . or dapr run -k -f dapr.yaml, the applications defined in the dapr.yaml Multi-App Run template file starts in Kubernetes default namespace.

Note: Currently, the Multi-App Run template can only start applications in the default Kubernetes namespace.

The necessary default service and deployment definitions for Kubernetes are generated within the .dapr/deploy folder for each app in the dapr.yaml template.

If the createService field is set to true in the dapr.yaml template for an app, then the service.yaml file is generated in the .dapr/deploy folder of the app.

Otherwise, only the deployment.yaml file is generated for each app that has the containerImage field set.

The files service.yaml and deployment.yaml are used to deploy the applications in default namespace in Kubernetes. This feature is specifically targeted only for running multiple apps in a dev/test environment in Kubernetes.

You can name the template file with any preferred name other than the default. For example:

  1. dapr run -k -f ./<your-preferred-file-name>.yaml

The following example includes some of the template properties you can customize for your applications. In the example, you can simultaneously launch 2 applications with app IDs of nodeapp and pythonapp.

  1. version: 1
  2. common:
  3. apps:
  4. - appID: nodeapp
  5. appDirPath: ./nodeapp/
  6. appPort: 3000
  7. containerImage: ghcr.io/dapr/samples/hello-k8s-node:latest
  8. createService: true
  9. env:
  10. APP_PORT: 3000
  11. - appID: pythonapp
  12. appDirPath: ./pythonapp/
  13. containerImage: ghcr.io/dapr/samples/hello-k8s-python:latest

Note:

  • If the containerImage field is not specified, dapr run -k -f produces an error.
  • The createService field defines a basic service in Kubernetes (ClusterIP or LoadBalancer) that targets the --app-port specified in the template. If createService isn’t specified, the application is not accessible from outside the cluster.

For a more in-depth example and explanation of the template properties, see Multi-app template.

Logs

The run template provides two log destination fields for each application and its associated daprd process:

  1. appLogDestination : This field configures the log destination for the application. The possible values are console, file and fileAndConsole. The default value is fileAndConsole where application logs are written to both console and to a file by default.

  2. daprdLogDestination : This field configures the log destination for the daprd process. The possible values are console, file and fileAndConsole. The default value is file where the daprd logs are written to a file by default.

Log file format

Logs for application and daprd are captured in separate files. These log files are created automatically under .dapr/logs directory under each application directory (appDirPath in the template). These log file names follow the pattern seen below:

  • <appID>_app_<timestamp>.log (file name format for app log)
  • <appID>_daprd_<timestamp>.log (file name format for daprd log)

Even if you’ve decided to rename your resources folder to something other than .dapr, the log files are written only to the .dapr/logs folder (created in the application directory).

Watch the demo

Watch this video for an overview on Multi-App Run in Kubernetes:

Next steps

Last modified March 21, 2024: Merge pull request #4082 from newbe36524/v1.13 (f4b0938)