github kubecost/cost-analyzer-helm-chart v1.95.0

latest releases: v2.5.0-rc.0, v2.4.2, v2.4.2-rc.1...
2 years ago

What Changed

Major Features

  • Migration to Single Page Application to increase performance and user experience.
  • GCP account-level tags
  • Inspecting resources now shows pod OOM and CPU throttling predictions
  • 371 Support OIDC authentication
  • Automatically retry failed ETL hourly builds
  • 1213 Internal Audit APIs to enhance Kubecost Algorithms (Publicly available later this year)

Minor

  • 1505 Require new installations of Kubecost to specify a cluster ID
  • 1465 Adds a no cache message when caches aren't available
  • 1481 Add k3s cluster to workflow
  • 1248 Introduced an API for diffing Assets between two windows
  • Improved accuracy of current cluster costs on the Cluster Right-sizing page
  • Improve performance of the Cluster Right-sizing page by querying a less granular datastore by default.
  • Improve error handling on Cluster Right-sizing API queries
  • 1269 Update KubeConfig Loader with Auth Import

Bugs

  • 1496 fix Grafana pod utilization dashboard query
  • Fix bugs in handling warnings in cluster sizing
  • 1280 Fix bug with Cloud Usage File Status
  • Return on config check
  • Sourcemaps upload now targeting the right release
  • Request Sizing Recommendations errors out with [object]
  • 1266 Fix GPU Bug
  • Add margin to account for tick labels in charts
  • 1393 Handle labels and aliased labels in savingsData cal
  • Values cut off on details.html graph
  • AWS marketplace costs filtered by Athena query

Open Source Community

  • 1279 Support GCP Spot VMs
  • 1457 Remove duplicate label configurations on service monitor

RC.1 Bug Fixes and Enhancements

  • 1293 Fixed prom querys and removed a leading space that was causing local cluster to not load from the file.
  • Support running SPA at a subpath URL and update redirects for action.html pages to target the new routes.
  • Improved bug reporting
  • Display ARM architecture only when available in cluster right-sizing
  • Scope asset query error outside of select statemen
  • Improved response notification when no instance types found for cluster sizing recommendations
  • 1513 Upgraded Prometheus from v2.31.1 to v2.35.0. This picked up golang v1.18 to address known security vulnerabilities.
  • 1300 Fixed a panic when querying allocations.
  • Moved license key or start a free trial to the top of the settings page
  • Cloud Integration card fix
  • Fix null check in cluster sizing recs
  • Alert after successful save

v1.95.0 Bug Fixes and Enhancements

  • If Allocation Files load, then do not attempt to compute allocation
  • Use context for model configs
  • fix aliased label bug
  • Adds the last 15 days to asset dropdown options
  • Make URLs sharable for first ime visitors
  • Add prom status to diagnostics
  • 1307 Update Azure LB pricing and comment
  • Update custom pricing settings
  • Fix Azure node pricing cache flow
  • Rebuilds the license management system on the FE to be more reliable
  • Savings page local disks card update
  • Fixes for read-only users related to product key, and GPU label support
  • Bug in the redirect logic
  • Improved the stability of Cluster Health Alerts.
  • Fixed broken node utilization page slider

Don't miss a new cost-analyzer-helm-chart release

NewReleases is sending notifications on new releases.