Who contributed to Kubernetes on July 12, 2026?

6 developers shipped this update, including Rajalakshmi-Girish, hakman, kubernetes-prow[bot], Roslaan001, minikube-bot, and SURAJ KUMAR.

What were the notable Kubernetes updates?

Fix kmsg watcher restart and shutdown issues, Merge pull request #1311 from hakman/kmsg-watcher-fixes, and Promote node-problem-detector v1.34.4, v1.35.3, v1.36.0.

@kubernetes

Container orchestration — what platform teams ship on

github ↗

Pick a date

Topics: Go Infrastructure Full archive →

The Wire · Showcase

NODE-PROBLEM-DETECTOR FIXES CRITICAL KMSG WATCHER DEADLOCK

By RepoJournal · Filed 06:56 UTC on July 12, 2026 · About Kubernetes

6 people shipped this

hakman @hakman 3 cited

Rajalakshmi-Girish @Rajalakshmi-Girish 1 cited

kubernetes-prow[bot] @kubernetes-prow[bot] 1 cited

Roslaan001 @Roslaan001 1 cited

minikube-bot @minikube-bot

SURAJ KUMAR @krsnaSuraj

A cascading restart bug in node-problem-detector's kmsg watcher caused 37k restarts in 200ms and could deadlock the shutdown path, and the fix just landed.

The kmsg watcher restart mechanism had three interlocking failure modes that only showed up under load [1]. The worst: a parser whose channel closes right after a successful restart triggered a hot restart loop with zero delay, measured at 37k restarts in 200 milliseconds. A second issue deadlocked Stop() when logCh filled up and the consumer was already shutting down, blocking watchLoop on a plain send. The third was a double-close on deferred cleanup when Stop() interrupted a restart attempt. All three are fixed in the same PR [2]. In parallel, node-problem-detector shipped three releases: v1.34.4, v1.35.3, and v1.36.0 [3], with v1.36.0 including the kubelet service name pattern update and dependency bumps [4]. The infrastructure team is also pivoting IBM ppc64le boskos resources from the deprecated wdc06 datacenter to the eu-de-1 region [5], which will improve test reliability for power-based architectures. Minikube landed a cleaner dependency consolidation, replacing golang.org/x/mod/semver with the already-present blang/semver library for Helm validation [6], plus routine addon updates across inspektor-gadget, kong, and cloud-spanner.

Action items

→ Update node-problem-detector to v1.36.0 before next cluster upgrade kubernetes/node-problem-detector [plan]
→ Review kmsg watcher logs for restart storms on existing deployments kubernetes/node-problem-detector [monitor]
→ Bump minikube dependency consolidation into next release kubernetes/minikube [plan]

References

[1] Fix kmsg watcher restart and shutdown issues ↗ kubernetes/node-problem-detector
[2] Merge pull request #1311 from hakman/kmsg-watcher-fixes kubernetes/node-problem-detector
[3] Promote node-problem-detector v1.34.4, v1.35.3, v1.36.0 ↗ kubernetes/k8s.io
[4] v1.36.0 ↗ kubernetes/node-problem-detector
[5] Replace wdc06 DC with eu-de-1 region in boskos on ibm ppc64le ↗ kubernetes/k8s.io
[6] addons: use blang/semver for helm version validation ↗ kubernetes/minikube

Quick answers

What shipped in Kubernetes on July 12, 2026?: A cascading restart bug in node-problem-detector's kmsg watcher caused 37k restarts in 200ms and could deadlock the shutdown path, and the fix just landed. In total, 43 commits, 21 pull requests, and 3 releases landed.
Who contributed to Kubernetes on July 12, 2026?: 6 developers shipped this update, including Rajalakshmi-Girish, hakman, kubernetes-prow[bot], Roslaan001, minikube-bot, and SURAJ KUMAR.
What were the notable Kubernetes updates?: Fix kmsg watcher restart and shutdown issues, Merge pull request #1311 from hakman/kmsg-watcher-fixes, and Promote node-problem-detector v1.34.4, v1.35.3, v1.36.0.

CADVISOR BUMPS PULL GOLANG ECOSYSTEM FORWARD, TEST INFRASTRUCTURE STABILIZES

Kubernetes core jumped cadvisor to v0.60.4, dragging golang.org/x/net from pseudo-versions into stable release territory and removing a long-standing protobuf dependency in one move.

+10

go 97 shipped 1-min read

@kubernetes 2 days ago

KOPS FIXES DNS-CONTROLLER ORPHAN BUG, PROW GAINS CLOUD METADATA INJECTION

Kops now automatically cleans up dns-controller deployments when clusters migrate away from gossip DNS, solving a manual deletion problem that's plagued operators for releases.

go 39 shipped 2-min read

@kubernetes 3 days ago

ETCD 3.7.0 GOES LIVE ACROSS THE STACK

Kubernetes is shipping etcd 3.7.0 final across test infrastructure, kops, and the main distribution, ending months of release-candidate testing and unlocking production etcd upgrades for 1.37+.

go 135 shipped 1-min read

@kubernetes 4 days ago

STRUCTURED MERGE REGRESSION FIXED ACROSS CONTROL PLANE

Kubernetes rolled back a breaking change in server-side apply that corrupted nullable container patches, with the fix propagating across kube-scheduler, endpoints, and resource allocation components within hours.

go 70 shipped 1-min read

— or follow the whole beat:

Elsewhere on the wire

Elixir & Phoenix about 5 hours ago

ELIXIR CLOSES COMPLETION AND OPERATOR SIGNATURE GAPS

The Elixir team shipped fixes for two long-standing IDE friction points: cursor completion after operators and proper type signatures for the ++ concatenation operator.

elixir 2 shipped 1-min read

Mobile platforms about 5 hours ago

SWIFT TIGHTENS AVAILABILITY DIAGNOSTICS AS FLUTTER TRACKS SKIA UPDATES

Swift's compiler now correctly diagnoses uses of unavailable API members in universally unavailable extensions, closing a gap that let invalid code slip through.

mobile 8 shipped 1-min read

@dotnet about 5 hours ago

ANDROID SSL VALIDATION BYPASS PATCHED; WASM JIT ALIGNMENT FIXES BLOCKER

SslStream on Android was ignoring platform certificate pinning entirely, bypassing network_security_config.xml and allowing rejected certificates through.

dotnet 28 shipped 1-min read

@spring-projects about 5 hours ago

SPRING-AI FIXES SILENT DATA LOSS IN BATCH EMBEDDING STRATEGY

A critical bug in TokenCountBatchingStrategy was dropping duplicate documents during batch processing, silently returning fewer embeddings than requested.

java 9 shipped 1-min read