Changes and Releases
Updates on Giant Swarm workload cluster releases, apps, UI improvements and documentation changes.
Added
- Add cleanup policy to remove old
trivy-operator resources.
Changed
- Enable
cleanup-controller with VericalPodAutoscaler by default. - Add missing ingress to
cleanup-controller CiliumNetworkPolicy. - Add
before-hook-creation delete-policy for upstream hooks.
Added
- Added annotation
helm.sh/resource-policy: keep on CRDs to prevent them from being pruned in an unexpected rollback event.
Changed
- Change “Node utilization” dashboard to use allocatable over capacity, to better reflect the available
resources of the nodes.
Changed
- Upgraded chart dependency from kube-prometheus-stack-58.5.2 to kube-prometheus-stack-58.5.2
- prometheus upgraded from 2.51.2 to 2.52.0
- thanos ruler upgraded from 0.34.1 to 0.35.0
Changed
- Upgraded chart dependency from kube-prometheus-stack-58.5.2 to kube-prometheus-stack-58.5.2
- prometheus upgraded from 2.51.2 to 2.52.0
- thanos ruler upgraded from 0.34.1 to 0.35.0
Fixed
- Fix loki and mimir mixins recording rules
- Fix atlas dashboard tags.
- Fix storage related panes on zot’s dashboards
- prometheus: scraping info can now be filtered by cluster
- add some basic linting configuration so we can track down issues in dashboards.
Added
- ServiceMonitors overview dashboard
- ServiceMonitors details dashboard
Added
- Support unified cluster-aws app. With cluster-aws v0.76.0 and newer, default apps are deployed with cluster-aws and default-apps-aws app is not deployed anymore.
Changed
- Default value for CAPA Node Pool
rootVolumeSizeGB was decreased from 300 to 8.
Changed
- Upgrade Gateway API CRDs to v1.1.0
Changed
- Add resources requests for the gateway.
- Add hpa resource and enable it by default for the querier and the distributor.
- Enable hpa for the gateway.
