Changes and Releases

Updates on Giant Swarm workload cluster releases, apps, UI improvements and documentation changes.

  • Changed

    • externalDNS annotations won’t no longer be set on the ingress services if baseDomain is not set. (#321)

    Removed

    • Default value for baseDomain configuration value automatically set for workload cluster installations. (#321)
    • Unused configuration values for chart installations on management clusters. (#321)
  • Changed

    • Upgrade to upstream image v1.7.3 which increases some hard-coded timeouts for certain ACME issuers (ZeroSSL and Sectigo) (#243)
    • Update kubectl container version to 1.24.2 (#243)
  • Fixed

    • Set loki config.auth_enabled to false by default.
    • Made default request timeout configurable via the happa ConfigMap (#3536)

    Full Changelog: https://github.com/giantswarm/happa/compare/v1.44.1...v1.44.2

  • Announcements

    • Important: the latest tag alias is being removed. Some users have reported issues using the latest tag on our hosted registries (Docker Hub, Quay, etc.). We advise against using latest tags and don’t use them ourselves, so this tag is not kept up to date. Please switch to using a tagged version. We will be removing the latest tag from our public registries in the near future to avoid confusion.

    Added

    • Add missing monitoring options in the Helm chart values.yaml.
    • Support sharding report metrics across multiple instances of the exporter.
    • Set runAsNonRoot and use RuntimeDefault seccomp profile.
    • Make replica count configurable in Helm values.
    • Add configurable tolerations to Helm values.
    • Reconcile and expose metrics for CISKubeBenchReport custom resources.
  • Added

    • Add flags --gcp-control-plane-sa-email and --gcp-control-plane-sa-scopes to template cluster that specify a Google Cloud Platform service account and its scopes to a cluster’s control plane nodes

    Removed

    • Removed giantswarm.io/cluster label from the default apps bundle and the App representing a CAPI cluster.
    • Add highlights for the week ending on 2022-06-10 (#1462)
  • Fixed

    • startup crashloop due to incorrect initialDelay settings.
    • Compatibility with Ingress v1 API
    • Links in documentation
    • Increase ingester probe initialDelay to 300 seconds

    Changed

    • Documentation minor updates and fixes
    • Documentation: how to test your Loki deployment
    • Update icon
  • Fixed

    • Fixed broken relative URLs in the README

    Changed

    • Upgrade to upstream image v1.7.2 (#204). This version completely removes cert-manager API versions v1alpha2, v1alpha3, and v1beta1. If you need to upgrade your resources, this document explains the process.
    • Update pytest-helm-charts to version 0.7.0 and adjust dependencies (#239)
    • Update kubectl container version to 1.24.1 (#204)