Changes and Releases
Updates on Giant Swarm workload cluster releases, apps, UI improvements and documentation changes.
Added
- Route Falco security events to the Giant Swarm Security Grafana organization. Operational data still goes to the default tenant.
- Enable metrics scraping for Falcosidekick and k8s-metacollector.
- Send Falco events to Falcosidekick by default.
Changed
- Enabled
json_output for Falco so security events land in Loki as structured JSON. - Allow additional properties for vendored upstream chart values so upstream keys are not rejected by the generated schema.
- Updated
falco to upstream version v0.44.1. - Updated
charts to upstream version falco-9.1.0.
Fixed
- Take the
app.kubernetes.io/version label from the chart app version instead of the chart version, as described in the Helm chart best practices. The chart version carries build metadata that pushed the label past the 63 byte limit and made the install fail.
Fixed
- Bump
kube-state-metrics sub-chart from 7.2.1 to 7.8.1, moving flux-ksm from kube-state-metrics v2.18.0 to v2.19.1. v2.18.0 leaks a reflector goroutine and its API server watch on every discovery poll cycle (kube-state-metrics#2867, fixed by #2920 in v2.19.0), which grows the workload cluster API servers until control plane nodes hit MemoryPressure.
Fixed
- (version) Report the release version instead of “dev” (pkg/project, stamped by CI and the Makefile) in #341 by @teemow
Full Changelog: https://github.com/giantswarm/klausctl/compare/v0.3.2...v0.3.3
Fixed
- Bump
kube-state-metrics sub-chart from 7.2.1 to 7.8.1, moving flux-ksm from kube-state-metrics v2.18.0 to v2.19.1. v2.18.0 leaks a reflector goroutine and its API server watch on every discovery poll cycle (kube-state-metrics#2867, fixed by #2920 in v2.19.0), which grows the workload cluster API servers until control plane nodes hit MemoryPressure.
Added
- (agent-platform) A session switcher rail on the session detail page, with session states derived server-side in #2302 by @marians
Full Changelog: https://github.com/giantswarm/backstage/compare/v0.236.0...v0.237.0
Added
- (clusters) Show node pool configuration alongside node metrics in #2262 by @marians
Full Changelog: https://github.com/giantswarm/backstage/compare/v0.235.1...v0.236.0
Fixed
- (gs) The installation inventory counts a connecting installation with a cached answer as still settling in #2301 by @teemow
Full Changelog: https://github.com/giantswarm/backstage/compare/v0.235.0...v0.235.1
Added
- (agent-platform) One installation scope for the section — home installation first, other installations as groups, installation on every row and detail page in #2300 by @teemow
Full Changelog: https://github.com/giantswarm/backstage/compare/v0.234.1...v0.235.0
Fixed
- (muster) The dashboard does not probe an installation the backend reports unreachable in #2299 by @teemow
Full Changelog: https://github.com/giantswarm/backstage/compare/v0.234.0...v0.234.1
