RepoJournal
Kubernetes

@kubernetes

Container orchestration — what platform teams ship on

Pick a date

  1. Tue 26 may
  2. Wed 27 may
  3. Thu 28 may
  4. Fri 29 may
  5. Sat 30 may
  6. Sun 31 may
  7. Mon 1 jun
  8. Mon 8 jun
  9. Tue 9 jun
  10. Wed 10 jun
  11. Thu 11 jun
  12. Fri 12 jun
  13. Sat 13 jun
  14. Sun 14 jun
  15. Mon 15 jun
  16. Tue 16 jun
  17. Wed 17 jun
  18. Thu 18 jun
  19. Fri 19 jun
  20. Sat 27 jun
  21. Sun 28 jun
  22. Mon 29 jun
  23. Tue 30 jun
  24. Wed 1 jul
  25. Thu 2 jul
  26. Fri 3 jul
  27. Sat 4 jul
  28. Sun 5 jul
  29. Mon 6 jul
  30. Today 7 jul

The Wire · Showcase

DRA DEVICE TAINTS GO GA, AUTOSCALER MULTI-ZONE FIX LANDS

By RepoJournal · Filed · About Kubernetes

Device Resource Allocation graduated to general availability overnight while kOps fixed a critical cluster-autoscaler blind spot that could strand scale-ups in single zones.

DRA Device Taints and Tolerations is now GA via the resource.k8s.io/v1 API [1], marking the feature ready for production workloads. The core team also cleaned up lingering dead code from the DRA resource slice tracker [2] [3], removing DeviceClass event processing that was causing unnecessary resyncs after the filtering logic was already removed. On the GCE side, kOps shipped a fix for multi-zone instance groups where cluster-autoscaler only saw the first zone's MIG, causing scale-ups to get stuck when that zone hit capacity constraints like spot shortages [4]. The fix registers each zonal MIG independently with autoscaler rather than treating them as a single node group. In validation land, CEL expressions in VAP and MAP now consistently use typed representations when available [5], fixing inconsistencies that only surfaced on informer cache misses. The declarative validation API tags also moved from alpha to beta [6], signaling stability for the feature.

Action items

References

  1. [1] DRA: Device Taints and Tolerations GA graduation ↗ kubernetes/kubernetes
  2. [2] DRA resourceslice tracker: ignore DeviceClass events kubernetes/kubernetes
  3. [3] DRA resourceslice tracker: remove DeviceClass informer option kubernetes/kubernetes
  4. [4] gce: register all zonal MIGs of a multi-zone instance group with cluster-autoscaler ↗ kubernetes/kops
  5. [5] Fix VAP/MAP params to consistently used the typed representation when available ↗ kubernetes/kubernetes
  6. [6] Promote declarative validation API tags from alpha to beta ↗ kubernetes/kubernetes

FAQ

What changed in Kubernetes on July 7, 2026?
Device Resource Allocation graduated to general availability overnight while kOps fixed a critical cluster-autoscaler blind spot that could strand scale-ups in single zones.
What should Kubernetes teams do about it?
Upgrade to latest kOps if running multi-zone GCE clusters with cluster-autoscaler enabled • Plan migration of DRA workloads from alpha to GA resource.k8s.io/v1 API • Review VAP/MAP CEL expressions if you've observed inconsistent typed vs unstructured behavior
Which Kubernetes repositories shipped on July 7, 2026?
kubernetes/kubernetes, kubernetes/kops

Related across the cluster

For your repos

The showcase is a teaser.
Your wire is the product.

Same engine. Different stack. Below: what changes when the wire is yours.

Showcase wire

  • •14 famous open source orgs
  • •One wire per day
  • •Public, generic
  • •Read on the web, when you remember

Your wire

  • →Up to 1,500 of your repos - orgs, deps, vendors
  • →Morning and evening briefs
  • →Action items routed to your team
  • →Slack delivery, email, breaking-news CVE alerts

Want a hands-on demo first? Ask a current user for an invite link.