Overview

This is an automated collection of upstream activity from github.

🔥 High Priority Updates

kubernetes/autoscaler#9764: [cluster-autoscaler-release-1.34] azure: implement AtomicIncreaseSize for VMSS

What type of PR is this?

/ki…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

kubernetes/autoscaler#9763: [cluster-autoscaler-release-1.35] azure: implement AtomicIncreaseSize for VMSS

What type of PR is this?

/ki…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

kubernetes/autoscaler#9762: test: mixed consumes/shared counters consumption

What type of PR is this?

###…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

kubernetes/autoscaler#9759: Add observed status fields to the CQ proposal

What type of PR is this?

/kind…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

kubernetes/autoscaler#9758: Feat: VPA InPlaceCapacityAware

What type of PR is this?

###…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: Yes

kubernetes/autoscaler#9757: AEP 9726 Capacity-Aware In-Place Updates

What type of PR is this?

/kind documentation

/kind…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

kubernetes/autoscaler#9751: Add CapacityQuota test case for nil selector

What type of PR is this?

/kind…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

kubernetes/enhancements#6165: KEP-6164: Add internal version elimination KEP

KEP issue: https://github.com/kubernetes/enhancements/issues/6164

This was discussed in the May 27, 2026 SIG API Machinery meeting where there was consensus to pursue this appro…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: undefined
  • State: open
  • Draft: No

Updates

kubernetes/kubernetes#139526: WatchList holds watchCache.RLock for 2x O(N) during initial events setup (vs) O(N) for equivalent List path, causing writer starvation at scale

What would you like to be added?

Reduce the watchCache.RLock critical section during WatchList initial events setup from 2x O(N) to O(N) if not O(1) (by leveraging the existing snapshot to build the event interval outside the lock) OR at the very least bring watch-list back in parity with…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: 4
  • State: open

kubernetes/kubernetes#139519: Windows KubeProxy fails to recreate LoadBalancer after HNS restart with error “hcnCreateLoadBalancer failed in Win32: Cannot create a file when that file already exists. (0xb7)“

What happened?

During HNS restart stress scenarios combined with rapid pod scale operations, Windows KubeProxy intermittently fails to recreate LoadBalancers and logs the following error:

hcnCreateLoadBalancer failed in Win32: Cannot create a file when that file already exists. (0xb7)

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: 4
  • State: open

kubernetes/kubernetes#139515: kube-proxy nftables: entire sync fails when any single Service generates a comment exceeding 128 characters

What happened?

When running kube-proxy in mode: nftables, if any Service causes kube-proxy to generate an nftables comment exceeding the Linux kernel’s 128-character limit, the entire nftables transaction is rejected and no service routing rules are applied on that sync cycle. kube-proxy …

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: 4
  • State: open

kubernetes/kubernetes#139513: kube-proxy nftables: sync time grows super-linearly with Service count

What happened?

Our production clusters run kube-proxy in ipvs mode. We evaluated migrating to nftables mode and observed that kube-proxy takes a long time to complete a sync, with the nft process at 100% CPU for most of that time.

This looks related to [#135639](https://github.com/kub

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: 3
  • State: open

kubernetes/kubernetes#139531: proxy/nftables: skip ListAll on incremental sync

What type of PR is this?

/kind bug

What this PR does / why we need it:

Refs #139513

Issue #139513 reports that kube-proxy nftables incremental sync time grows super-linearly with Service count, and that at 44k Services the incremental sync is almost entirely spent in `nft —json —ters…

🔗 Link

Metadata:

  • Created: 2026-06-06
  • Comments: undefined
  • State: open
  • Draft: Yes

kubernetes/kubernetes#139530: Set short termination grace period for test pods in MemoryQoS tests

What would you like to be added: I would like to add new metrics:

  • kube_pod_container_status_resource_requests (Gauge) - The number of actual requested request resources currently allocated to a container (status.containerS…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: 1
  • State: open

kubernetes/kube-state-metrics#2979: Surface DRA ResourceClaim references in the pod collector (kube_pod_resource_claim_info)

What would you like to be added:

A new ALPHA metric on the existing pod collector that surfaces a Pod’s Dynamic Resource Allocation (DRA) ResourceClaim references, derived solely from the Pod object:

  • kube_pod_resource_claim_info — one series per entry in pod.spec.resourceClaims[], with…

🔗 Link

Metadata:

  • Created: 2026-06-05
  • Comments: 1
  • State: open

prometheus/alertmanager: 0.32.2 / 2026-06-05

  • [BUGFIX] Fix dispatcher goroutine leaks on destroyed alertgroup swap. #5241

🔗 Link

Metadata:

  • Version: v0.32.2
  • Published: 2026-06-05
  • Prerelease: No

envoyproxy/gateway: v1.8.1

Release Announcement

Check out the v1.8.1 release announcement to learn more about the release.

What’s Changed

🔗 Link

Metadata:

  • Version: v1.8.1
  • Published: 2026-06-05
  • Prerelease: No

envoyproxy/gateway: v1.7.4

Release Announcement

Check out the v1.7.4 release announcement to learn more about the release.

What’s Changed

🔗 Link

Metadata:

  • Version: v1.7.4
  • Published: 2026-06-05
  • Prerelease: No

containerd/overlaybd: v1.0.17

Features

  • add live snapshot API with device ID tracking and layer restacking #387 (Yifan Yuan)
  • adding https mode to credentialConfig #397 (Julius)

Bug Fixes

  • cache: cap actualEvict by totalUsed_ to…

🔗 Link

Metadata:

  • Version: v1.0.17
  • Published: 2026-06-05
  • Prerelease: No

containerd/overlaybd: Development Build

Bug Fixes

  • cache: cap actualEvict by totalUsed_ to prevent eviction deadloop #400 (yuchen.cc)
  • pass node label value to OverlayBD_Alive metric render #401 (Sri Harsha)

Continuous Integration

  • declare …

🔗 Link

Metadata:

  • Version: latest
  • Published: 2026-06-05
  • Prerelease: Yes

This content was automatically collected on 2026-06-06 02:55:06