Skip to content
This repository has been archived by the owner on Oct 25, 2022. It is now read-only.

Latest commit

 

History

History
49 lines (37 loc) · 4.28 KB

HowToUse.md

File metadata and controls

49 lines (37 loc) · 4.28 KB
Raw

Kyma VSCode Plugin

Installation

  • Clone the Github repository.
  • npm install
  • either use visual studio extension manager npm install -g vsce and vsce package to create a .vsix package or click F5 to run the extension on extension development host.
  • You can also download the vsix file directly from Github releases.

Configuration

Additional to kubernetes extension's own configuration settings, this extension only ask for a port to serve to schema validator file. This port is default to 4402.

  • Most important configuration is the kubeconfig file that is used to connect to Kyma instance. Provide an absolute path for it.

Functionality

Kyma Explorer

This plugin adds a new explorer view to Activity bar. It has the kyma logo and clicking it connects you to kyma instance. It shows the kyma deployments in a tree view. This tree view shows kubernetes resources as well as custom defined resources.

  • Right Clicking on a namespace and selecting Set as current namespace changes the current active namespace. It renders the tree view again.
  • Clicking on resources open their deployment yaml files. Very useful for inspecting.
  • This yaml files can be edited on the fly and be re-deployed.

Commands

Kyma commands use the command category kyma. Writing kyma to command pallette brings up all Kyma related commands.

  • Add New Environment : Adds a new environment (namespace) to kyma instance. In a new deployment, Kyma has qa , production , stage environments out of the box.

  • Delete Environment: Deletes the selected environment with all it's deployments.

  • Open Console: Opens console.kyma.local in local web browser.

  • Deploy to Kyma: If selected file is a javascript file, it tries to deploy it as a lambda to Kyma. It creates a deployment.yaml file with default settings and uses javascript file's name as identifier. Deployment is done to currently active namespace (Which can be seen from Kyma View). If the file is a yaml file, it creates an extra yaml file with the same content in temp folder and deploys it. The reason for creating temp files is the ability to re-deploy configuration files that can be seen from the kyma View.

  • Expose Lambda: Creates a default Api configuration for the lambda function and deploys it. You may need to modify your hosts file to reach the api.

  • Debug Lambda: Debugs the lambda function. If opened file is a javascript file, it takes the filename as the function name. If it's a yaml file, it takes name in metadata config. If none of the above, it asks the Kyma instance for function names and calls the one selected.

    Once a function is selected ,it sends a request to function using kubeless. If kubeless command exits normally, the output of the function is shown in OUTPUT channel named kyma lambda. If kubeless functions exits with an error code, that means there is an error in the function and thus closer inspection begins. The last 10 seconds of the logs of the pod that runs the lambda function is taken , words failed and error searched in the logs. If found, extension tries to parse the text for relevant line information. It is also shown in the PROBLEMS tab with the relevant error message. Most of this process is manual , so debugging couple of times might help identifying errors better.

Scheme Validation

This extension provides schema validation and hover support for multiple yaml files. If the yaml file has a key named ApiVersion, its value is parsed and according to yaml version a validation scheme is called. Supported versions are:

  • gateway.kyma.cx/v1alpha2
  • kubeless.io/v1beta1
  • servicecatalog.kyma.cx/v1alpha1
  • remoteenvironment.kyma.cx/v1apha1

Another hover support is, if you are in a yaml file except kubeless, when you hover to function key, you will see currently deployed lambda functions. This functionality is useful if you want to deploy an API for a function and want to get the function name easily just by hovering.

Dependencies

This extension needs both kubeless and kubectl in the path. This program relies on a fixed kubeless issue, so you may need to change the executable. For more info see this github issue.