Release v1.1.0 (v1.1.1)

[gianlucam76]

🚀 Features

• Sveltos MCP Server: This release introduces the new Sveltos MCP Server, which provides the ability to trigger debug sessions. This new component enhances the management capabilities of the Sveltos platform by providing a powerful way to interact with and diagnose cluster issues, particularly when used in conjunction with AI-based SRE tools.

• Sveltos Profile Management: You can now pause and unpause a Sveltos profile instance. When a profile is paused, Sveltos will stop reconciling it, meaning any changes to the profile or its matching clusters will not be propagated. This is useful for preventing unwanted changes during maintenance or troubleshooting. PR

• Predictable ClusterProfile Names: Event-driven ClusterProfile instances no longer have random names. A new, optional field, InstantiatedProfileNameFormat, has been added to the EventTrigger spec. This allows you to define a naming template for generated profiles, making it easy to predict their names and set up dependencies for other profiles. PR

• CEL-based Health Checks: You can now use Common Expression Language (CEL) to define ValidateHealths for resources. This provides more flexibility and power for creating custom health validation rules. PR

• Kustomize Components Support: ClusterProfile's KustomizationRef section now includes a components field. This lets you include reusable Kustomize components in your deployments by specifying relative paths to component directories. Sveltos will automatically integrate these components into the kustomization.yaml file before building and applying the final manifest. PR

• Enhanced Addon Reporting: The Show addons command now also reports on resources deployed directly in the management cluster, providing a more complete overview of your deployed components.

🐛 Bug Fixes

• Fixed an issue where the Helm chart was not correctly upgrading Sveltos Custom Resource Definitions (CRDs).

• Improved error messages for resource deployment failures. The messages now specify the exact resource that failed to deploy, instead of providing a general error message.

• Conflict messages have been improved to provide clearer information during reconciliation.

Heads up! ⚠️

Any ClusterProfile or Profile that uses Helm options will be redeployed after this upgrade. This is due to new options being added, which changes the profile's hash. ClusterProfile/Profile with Helm charts without any options will not be affected.