Cloud Functions Configuration

Kyvos now supports for configuring pre-created functions to enable cluster scheduling in Kyvos for new and post-deployments.

For AWS, pre-created Lambda functions are supported.
For Azure, pre-created Linux-based Function Apps are supported.

Points to know:

Support has been added to configure pre-created functions during wizard-based deployments, enabling scheduling to be set up immediately after the cluster deployment is complete.
Additionally, support is now available to configure pre-created functions post-deployment. This enables scheduling in existing cloud-based deployments where it was previously unavailable due to the unavailability of functions during the original deployment.
Functions are specified using only their function names, regardless of how they were created.
Permissions:
- For AWS, automated and wizard-based deployment, the following permissions are required:
  - "lambda:ListFunctions": It is used to list the functions.
  - "lambda:GetPolicy": It is used to create the function URL and event ID.
- For Azure, no additional permission except contributor is required.
- For GCP wizard-based and post-deployment, the following permission is required:
  - compute.regions.list
Create a Linux-based function app.

Create a Linux-based function app for Azure

For Azure, Function App in a different resource group is not supported. Ensure that you must create the function App in the same resource group as your Kyvos Deployment.

You can create a Linux-based function app using ARM templates or Terraform script.

CASE 1: Using ARM template

Note

You can create secure or non-secured function app as needed. To create a secured function app, download the ‘functions_secure.json’ file. Similarly, for non-secured, download the ‘functions.json’ file from the Azure Installation Files folder.

Login to Azure portal.
Search for Deploy a custom template.
Upload the ARM template and save it. The Custom Deployment page is displayed.
Complete on-screen parameters. Click the i icon for more information about on-screen parameters.

CASE 2: Using Terraform script

Download the functions.tf script and terraform.tfvars parameter file.
Complete all parameters in the terraform.tfvars file.
Run the following commands in the same sequence in the folder where the script and parameter files are located.
1. terraform init
2. terraform plan
3. terraform apply
If Terraform is not installed in your local system, you can use Azure Cloud Shell to execute the above-mentioned commands.

To configure the cloud functions, perform the following steps.

On the navigation pane, click Kyvos and Ecosystem > Cloud Functions. The Cloud Functions page is displayed. The details on the page vary depending on the type of cloud.
Select the functions from the list for the following:
- For AWS, specify the following:
  1. Cluster Scaling Schedule
  2. Cluster Status
  3. BI Server Start
- For Azure, specify the following:
  1. Select the required Function App from the list.
Click Save. The functions are now configured to enable cluster scheduling in Kyvos. You can now use cluster scheduling.