Kubernetes Releases¶
Nubenetes V2 Elite Portal
You are browsing the AI-Curated V2 Elite Edition. Looking for the exhaustive list of references? Check out the V1 Historical Archive.
Architectural Context
Detailed reference for Kubernetes Releases in the context of The Container Stack.
Cloud Integrations¶
AWS¶
Security¶
- (2023) Private Access to the AWS Management Console is generally available [COMMUNITY-TOOL] โ Details the general availability of Private Access to the AWS Management Console. This feature allows enterprise customers to limit console access to connections originating from specific network endpoints or VPCs, drastically reducing potential attack surfaces and ensuring secure administration paths.
Cloud Native¶
Kubernetes¶
Internals¶
- (2026) Kubernetes v1.37: etcd RangeStream Cuts Memory Use on Large List Reads [EN CONTENT] [ADVANCED LEVEL] ๐๐๐๐ [DE FACTO STANDARD] [ENTERPRISE-STABLE] [ETCD] [KUBERNETES] [PERFORMANCE] [SRE] โ The RangeStream feature in etcd transitions to beta in Kubernetes v1.37, addressing the historical architectural bottleneck of memory exhaustion during massive, unpaginated
LISToperations. By implementing native server-side streaming for range requests, it drastically reduces the peak memory footprint across both the API server and underlying etcd storage nodes. While requiring updated client-side configurations to maximize throughput, this optimization is foundational for scaling volatile 2026 AI scheduling workloads and establishes a highly resilient baseline for control plane Day-2 High Availability.
Release Management¶
- (2026) Kubernetes v1.37 Sneak Peek [EN CONTENT] ๐๐๐๐ [ARCHITECTURE] [DE FACTO STANDARD] [DEPRECATIONS] [ENTERPRISE-STABLE] [GPU] [RELEASE] โ The Kubernetes v1.37 release sneak peek outlines critical API transitions, deprecations, and core architectural shifts necessary for maintaining Day-2 cluster health. From a Curator Insight perspective, early visibility into removed features allows SREs to proactively refactor Helm charts and Operators before control plane upgrades. Live grounding confirms that this release cycle heavily matures Dynamic Resource Allocation (DRA) for GPU-intensive AI workloads and solidifies Gateway API implementations, acting as a mandatory preparation guide for platform engineering teams.
Releases¶
- (2026) Kubernetes v1.37: Memory QoS Graduates to Beta [EN CONTENT] ๐๐๐๐ [CGROUPV2] [DE FACTO STANDARD] [ENTERPRISE-STABLE] [KUBERNETES] [LEGACY] [LLM] [QOS] [SRE] โ Memory QoS has reached Beta in Kubernetes v1.37, fundamentally shifting node-level resource management by natively leveraging Linux cgroup v2 to enforce sophisticated memory throttling instead of relying on abrupt OOM kills. This architectural evolution allows Site Reliability Engineers to maximize node utilization for spiky workloads, such as LLM inference and data processing pipelines, without jeopardizing the stability of co-located agents or system daemons. While enabled by default, production clusters must ensure underlying node OS compatibility with cgroup v2, transitioning legacy memory limit strategies toward dynamic, intent-based Quality of Service parameters.
- (2026) Kubernetes v1.37: Introducing Node Lifecycle Conditions [EN CONTENT] ๐๐๐๐ [ARCHITECTURE] [DE FACTO STANDARD] [ENTERPRISE-STABLE] [GPU] [KUBERNETES] [LEGACY] [NODE-MANAGEMENT] [SRE] โ Kubernetes v1.37 introduces Node Lifecycle Conditions, providing a standardized, highly granular state machine for node health that consolidates legacy Readiness, Taint, and Pod state mechanisms. From an architectural curation perspective, this shift empowers SRE teams to construct more deterministic auto-remediation and workload eviction pipelines during Day-2 operations. Grounding this feature in the current cloud-native ecosystem, it natively enhances Dynamic Resource Allocation (DRA) awareness, serving as a critical infrastructure primitive for maintaining high availability in GPU-heavy and agentic AI clusters.
- (2026) Kubernetes v1.37: Garhwal [EN CONTENT] ๐๐๐๐ [ARCHITECTURE] [DE FACTO STANDARD] [ENTERPRISE-STABLE] [RELEASE] [UPGRADE] โ Official announcement for Kubernetes v1.37, codenamed Garhwal. This release introduces significant API stabilizations and security primitives, emphasizing control plane resilience and enhanced operator capabilities. Serves as the definitive reference for cluster administrators planning enterprise upgrade cycles.
Resource Allocation¶
- (2026) Kubernetes v1.37: DRA Updates [EN CONTENT] [ADVANCED LEVEL] ๐๐๐๐ [DE FACTO STANDARD] [DRA] [ENTERPRISE-STABLE] [GPU] [KUBERNETES] [LEGACY] [LLM] โ Curator Insight: Kubernetes v1.37 fundamentally modernizes heterogeneous resource management, elevating Dynamic Resource Allocation (DRA) to a production-grade standard for enterprise AI platforms. Live Grounding: Official release telemetry confirms that DRA's extended resource support allows operators to transition GPU workloads incrementally without modifying existing Pod manifests.
- Architectural Role: Replaces brittle legacy device plugins with a declarative, API-driven scheduling model engineered specifically for allocating scarce TPUs and accelerators across agentic LLM workloads.
- Production Trade-offs: While DRA drastically improves hardware utilization, platform SREs must rigorously enforce
ResourceQuotaboundaries, as cross-namespace hardware contention resolves on a strict first-come, first-served basis without native capacity borrowing.
Scheduling¶
- (2026) Kubernetes v1.37: Advancing Workload-Aware Scheduling [EN CONTENT] [ADVANCED LEVEL] ๐๐๐๐ [AI] [DRA] [EMERGING] [ENTERPRISE-STABLE] [GPU] [KUBERNETES] [LEGACY] [LLM] [MLOPS] [SCHEDULING] โ The release of Kubernetes v1.37 marks a definitive architectural shift from pod-centric placement to holistic workload-aware scheduling, specifically targeting the complexities of distributed AI/ML training and batch processing. Live web grounding confirms that foundational primitivesโsuch as Dynamic Resource Allocation (DRA) Extended Resources,
PodGroup, and the newCompositePodGroupAPIโhave matured, enabling hierarchical gang scheduling and native GPU topology constraints directly within the upstream scheduler. While this native integration reduces the need for external batch schedulers like Volcano, it introduces a transition cost, requiring platform teams to adopt new workload APIs and update legacy deployment patterns. Ultimately, by natively combining HPA scale-to-zero capabilities with advanced DRA device taints, v1.37 establishes a highly efficient, enterprise-stable control plane for modern agentic and LLM inference workloads.
Kubernetes (1)¶
Control Plane Architecture¶
API Server¶
Mixed Version Proxy¶
- (2026) Kubernetes v1.36: Mixed Version Proxy Graduates to Beta [EN CONTENT] [ADVANCED LEVEL] ๐๐๐๐ [EMERGING] [ENTERPRISE-STABLE] โ Architectural breakdown of the Mixed Version Proxy (KEP-4020) graduating to Beta and becoming enabled by default in Kubernetes 1.36. The mechanism prevents false 404 errors during mixed-version control plane rolling upgrades by proxying unrecognized resource requests to a newer peer API server. Enforces peer-aggregated discovery data flows, allowing operators to build and upgrade highly available control planes with zero client disruption.
Cloud Controller Manager CCM¶
Instrumentation¶
- (2026) Kubernetes v1.36: New Metric for Route Sync in the Cloud Controller Manager [EN CONTENT] [ADVANCED LEVEL] ๐๐๐๐ [EMERGING] [ENTERPRISE-STABLE] โ Analysis of the new alpha metric 'route_controller_route_sync_total' introduced in the Kubernetes v1.36 Cloud Controller Manager (CCM). Designed to facilitate A/B validation of the 'CloudControllerManagerWatchBasedRoutesReconciliation' feature gate, transitioning cloud routing loops from fixed polling intervals to reactive, watch-based reconciliation. Drastically reduces redundant control plane queries to underlying hyperscaler APIs, preventing rate-limiting blocks in large-scale cluster architectures.
Kubernetes Core¶
API Evolution¶
V1.16¶
- (2019) Kubernetes v1.16 API deprecation testing [SHELL CONTENT] ๐๐ [LEGACY] โ Curator Insight features a test script for identifying deprecated Kubernetes v1.16 APIs. Live engineering truth in 2026 finds this script outdated, though historically valuable for understanding the mechanics of migrating extensions/v1beta1 API resources.
V1.22¶
- (2021) kubernetes.io: Kubernetes API and Feature Removals In 1.22: Hereโs What You Need To Know [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight points to the critical official guide regarding upcoming 1.22 API removals. In 2026, this remains the historical checklist template used by platform engineers to design deprecation checklists.
V1.25¶
- (2022) kubernetes.io: Kubernetes Removals and Major Changes In 1.25 [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight highlights the upcoming removals in 1.25. Live grounding in 2026 registers 1.25 as the ultimate removal point for PodSecurityPolicy (PSP), finalizing the modern security validation ecosystem.
Cluster Security¶
V1.21¶
- (2021) devclass.com: Kubernetes 1.21 unloads pod security, adds dual IPv4/IPv6 networking, and shuts down gracefully [COMMUNITY-TOOL] โ Curator Insight discusses Kubernetes 1.21 dropping PodSecurityPolicy, adding dual IPv4/IPv6 networking, and implementing graceful shutdowns. In 2026, dual-stack networking is highly standard, having originated from these architectural changes.
V1.23¶
- (2021) blog.aquasec.com: Kubernetes Version 1.23: What's New for Security? [LEGACY] โ Curator Insight duplicates security focuses in 1.23. In 2026, it remains useful for security compliance audits when checking clusters against legacy security postures.
Container Runtimes¶
Dockershim Deprecation¶
- (2021) kubernetes.io: Dockershim removal is coming. Are you ready? [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight focuses on Dockershim removal readiness. In 2026, this remains a valuable reference artifact for migrating air-gapped systems or old enterprise clusters from Docker to containerd.
- (2020) zdnet.com: Kubernetes dropping Docker is not that big of a deal [COMMUNITY-TOOL] โ Curator Insight notes ZDNet's perspective that removing Dockershim is not a disruptive disaster. Live grounding in 2026 confirms this prediction was correct, as the ecosystem successfully migrated to containerd and CRI-O without massive developer friction.
- (2020) openshift.com: Kubernetes is Removing Docker Support, Kubernetes is Not Removing Docker Support [COMMUNITY-TOOL] โ Curator Insight identifies Red Hat's explanatory guide on the nuance of Docker support removal. Live grounding in 2026 verifies this piece as a classic, helping teams comprehend that standard OCI-compliant Docker containers remained compatible despite runtime changes.
V1.24¶
- (2022) infoq.com: Kubernetes Proceeding with Deprecation of Dockershim in Upcoming 1.24 Release [COMMUNITY-TOOL] โ Curator Insight outlines InfoQ's coverage of Dockershim's imminent demise in 1.24. Live grounding in 2026 records this as the definitive end of direct Docker Engine support within the kubelet codebase.
- (2022) kubernetes.io: Updated: Dockershim Removal FAQ [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight provides the updated Dockershim FAQ. In 2026, it serves as an excellent reference troubleshooting tool for understanding CRI-dockerd shims and standard CRI integrations.
Infrastructure¶
Registry¶
- (2022) kubernetes.io: registry.k8s.io: faster, cheaper and Generally Available (GA) [MARKDOWN CONTENT] [COMMUNITY-TOOL] โ Announces the transition of the primary Kubernetes container registry from 'k8s.gcr.io' to the community-owned, CDN-backed 'registry.k8s.io'. This infrastructural pivot reduces data egress costs and latency by intelligently routing traffic to regional cloud storage providers near the client nodes. Engineers must audit existing configuration files and pipelines to refer to the new endpoint to avoid build failures.
Job Scheduling¶
V1.21 (1)¶
- (2021) kubernetes.io: kubernetes 1.21: CronJob Reaches GA [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight provides the official blog announcement of CronJob reaching GA in Kubernetes 1.21. Live analysis in 2026 acknowledges this as the milestone that finalized the performance and controller upgrades for batch scheduling.
- (2021) kubernetes.io: Introducing Suspended Jobs in Kubernetes 1.21 [DOCUMENTATION] [COMMUNITY-TOOL] [MLOPS] โ Curator Insight introduces suspended jobs in 1.21. Live grounding in 2026 indicates this feature was a crucial building block for modern external batch orchestrators like Kueue to control execution pauses.
Node Management¶
Fault Tolerance¶
- (2022) kubernetes.io: Kubernetes 1.26: Non-Graceful Node Shutdown Moves to Beta [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Introduces the promotion of Non-Graceful Node Shutdown capabilities to Beta in Kubernetes 1.26. This feature enables stateful workloads (such as databases on StatefulSets) to recover gracefully on active nodes when a cluster host crashes abruptly without a proper shutdown path. This addresses a common issue where volume attachments remained locked to failed instances indefinitely.
V1.22 (1)¶
- (2021) kubernetes.io: Graceful Node Shutdown Goes Beta [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight highlights Graceful Node Shutdown transitioning to Beta in 1.22. Live grounding in 2026 confirms this functionality prevents database corruption and packet drops on spot instances by coordinating termination steps with systemd.
Observability¶
V1.21 (2)¶
- (2021) kubernetes.io: Kubernetes 1.21: Metrics Stability hits GA [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight records the GA of Metrics Stability in 1.21. Live analysis in 2026 confirms this stabilization framework remains critical for maintaining backwards-compatible prometheus exporters without breaking monitoring dashboards.
Operator Framework¶
V1.20¶
- (2020) thenewstack.io: Kubernetes 1.20 Enhances the Operator Experience and Brings New Features to the Container Runtime [COMMUNITY-TOOL] โ Curator Insight highlights the operator and container runtime interface advancements in 1.20. Live engineering review in 2026 uses this context to study early attempts to standardize execution metrics for operational tooling.
Release Lifecycle¶
Documentation¶
- (2026) relnotes.k8s.io: Kubernetes Release Notes [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight highlights the official Kubernetes Release Notes portal. Live grounding in 2026 demonstrates this tool's constant utility for DevOps engineers running automated deprecation pipelines and tracking version-to-version API mutations.
Process¶
- (2021) kubernetes.io: Kubernetes Release Cadence Change: Hereโs What You Need To Know [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight addresses the Kubernetes Release Cadence shift from 4 to 3 releases per year. Live grounding in 2026 confirms this release model has brought predictability and stability to platform teams everywhere.
V1.20 (1)¶
- (2020) thenewstack.io: Kubernetes 1.20 Lands with 44 Enhancements [COMMUNITY-TOOL] โ Curator Insight focuses on the 44 enhancements landing with Kubernetes 1.20. In 2026, this resource provides historic data on the stabilization of volume snapshots, PID limits, and early steps toward API deprecation tracking.
V1.21 (3)¶
- (2021) devopscube.com: Kubernetes v1.21 Released: Here is What you should know [COMMUNITY-TOOL] โ Curator Insight offers an overview of Kubernetes 1.21. In 2026, we track this release as a major evolution that brought CronJobs to GA and introduced PodDisruptionBudget stability.
- (2021) kubernetes.io: Kubernetes 1.21: Power to the Community [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight presents the official 1.21 release announcement 'Power to the Community'. From a 2026 perspective, 1.21 initiated the long-awaited deprecation of PodSecurityPolicy, shifting the community towards Pod Security Standards.
- (2021) openshift.com: Kubernetes 1.21 Grows Innovative New Features [COMMUNITY-TOOL] โ Curator Insight presents Red Hat's analysis of 1.21 features. Live grounding in 2026 reviews this for enterprise adaptation paths, specifically outlining how OpenShift clients prepared for eventual PSP deprecations.
V1.22 (2)¶
- (2021) sysdig.com: Kubernetes 1.22 โ Whatโs new? [LEGACY] โ Curator Insight highlights Sysdig's review of Kubernetes 1.22. Live grounding in 2026 views 1.22 as the milestone where major legacy APIs (like v1beta1 Ingress) were fully removed, causing teams to aggressively update Helm charts.
- (2021) kubernetes.io: Kubernetes 1.22: Reaching New Peaks [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight presents the 1.22 release announcement 'Reaching New Peaks'. In 2026, this represents the definitive documentation on client credential plugins and early server-side apply stabilization.
- (2021) thenewstack.io: Less Is More with Kubernetes 1.22 [COMMUNITY-TOOL] โ Curator Insight analyzes the 'Less Is More' paradigm of 1.22. Live grounding in 2026 reflects on how API cleanliness made the control plane leaner, decreasing memory footprint and etcd payload overhead.
- (2021) acloudguru.com: Whatโs new with Kubernetes 1.22? [COMMUNITY-TOOL] โ Curator Insight points to Pluralsight's educational summary of 1.22. In 2026, it serves as a baseline tutorial explaining the transition to cgroups v2, which has since become standard in all production distros.
V1.23 (1)¶
- (2021) sysdig.com: Kubernetes 1.23 โ Whatโs new? [COMMUNITY-TOOL] โ Curator Insight presents Sysdig's assessment of Kubernetes 1.23. Live review in 2026 shows this release solidified Dual-stack IPv4/IPv6 services to GA and introduced Ephemeral Volumes.
- (2021) thenewstack.io: Kubernetes 1.23: Dual Stack IPv4/IPv6, CronJobs, Ephemeral Volumes [COMMUNITY-TOOL] โ Curator Insight summarizes the GA of Dual-stack and CronJobs in 1.23. Live engineering analysis in 2026 verifies these features are highly mature, having stabilized completely during the 1.23 release cycle.
- (2021) kubernetes.io: Kubernetes 1.23: The Next Frontier [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight highlights the official Kubernetes 1.23 announcement 'The Next Frontier'. In 2026, this announcement marks the stabilization of ephemeral containers for interactive debugging.
V1.24 (1)¶
- (2022) sysdig.com: Kubernetes 1.24 โ Whatโs new? [COMMUNITY-TOOL] โ Curator Insight presents Sysdig's review of 1.24. Live analysis in 2026 reminds us that 1.24 removed Dockershim, introduced Beta volume expansion, and made gRPC probes standard.
V1.25 (1)¶
- (2022) sysdig.com: Kubernetes 1.25 โ Whatโs new? [EMERGING] โ Curator Insight reviews what is new in 1.25. In 2026, this is remembered for bringing Pod Security Admission to GA, KMS v2 alpha, and cgroups v2 enhancements.
Releases (1)¶
V1.25 (2)¶
- (2022) kubernetes.io: Kubernetes v1.25: Combiner [GO CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Kubernetes v1.25, dubbed 'Combiner', represents a major milestone in core stability and API maturation. The release is architecturally significant for the formal removal of PodSecurityPolicy (PSP) in favor of Pod Security Admission (PSA) and the promotion of several storage features to General Availability. Operators must adapt to these security and storage modifications to ensure workload compatibility.
V1.26 Analysis¶
- (2022) armosec.io: Kubernetes Version 1.26: Everything You Should Know [COMMUNITY-TOOL] โ A security-focused architectural assessment of Kubernetes 1.26. The paper highlights the deprecation of non-CSI storage plugins, the expansion of dynamic security standards, and the introduction of CEL validations in admission controllers. This analysis provides key insights for security engineers seeking to implement finer control policies directly within API servers.
V1.26 Features¶
- (2022) sysdig.com: Kubernetes 1.26 โ Whatโs new? [COMMUNITY-TOOL] โ A detailed technical review of the architectural enhancements shipped in Kubernetes 1.26. Key features highlighted include Common Expression Language (CEL) validation in CustomResourceDefinitions (CRDs), improvements to Pod Scheduling Readiness, and container registry migrations. This summary acts as an essential checklist for systems engineering teams coordinating scheduled cluster upgrades.
V1.27¶
- (2023) kubernetes.io: Kubernetes v1.27: Chill Vibes [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ The official documentation for the Kubernetes 1.27 release ('Chill Vibes'). Featuring 40 enhancement proposals, this release emphasizes security posture hardening, scheduling optimizations, and resource isolation. Prominent updates include the graduation of SeccompDefault to GA and the introduction of in-place container resource resizing.
V1.27 Features¶
- (2023) sysdig.com: Kubernetes 1.27 โ Whatโs new? [COMMUNITY-TOOL] โ An analytical overview of Kubernetes 1.27's flagship updates. Features evaluated include the stabilization of SeccompDefault, container image signature verification workflows, and advancements in dynamic resource allocation. This serves as a vital blueprint for enterprise platform teams coordinating upcoming cluster upgrades.
V1.27 Industry¶
- (2023) thenewstack.io: Kubernetes 1.27 Arrives [COMMUNITY-TOOL] โ Industry reporting on the release of Kubernetes 1.27, detailing the major developer and operator benefits. It focuses on how security stabilization, API cleanup, and enhanced cluster control interfaces enable enterprise systems to scale more safely and cost-effectively.
V1.27 Security¶
- (2023) armosec.io: Kubernetes 1.27 Release: Enhancements and Security Updates [COMMUNITY-TOOL] โ An in-depth security analysis detailing the security updates and modifications introduced in Kubernetes 1.27. It discusses shifts in default container security profiles, RBAC configurations, and validation tools, providing actionable guidelines for compliance and cluster hardening.
V1.28 Features¶
- (2023) thenewstack.io: Kubernetes 1.28 Accommodates the Service Mesh, Sudden Outages [COMMUNITY-TOOL] โ An industry report covering Kubernetes 1.28's key updates, notably native support for sidecar container lifecycles to aid service mesh deployments, and automated recovery controls for sudden node failures. These updates greatly reduce the coordination code needed for robust, multi-tenant microservices clusters.
Resource Allocation (1)¶
V1.21 (4)¶
- (2021) thenewstack.io: Kubernetes 1.21 Brings a New Memory Manager, More Flexible Scheduling [COMMUNITY-TOOL] โ Curator Insight highlights the introduction of a new Memory Manager and flexible scheduling in 1.21. Live grounding in 2026 shows these scheduling frameworks laid the groundwork for modern NUMA-aware multi-core resource allocation in HPC/AI setups.
V1.22 (3)¶
- (2021) kubernetes.io: Kubernetes Memory Manager moves to beta [DOCUMENTATION] [COMMUNITY-TOOL] โ Curator Insight explains the Kubernetes Memory Manager moving to Beta. Live engineering in 2026 points to this as the genesis of modern NUMA alignment for performance-sensitive low-latency microservices.
Resource Management¶
Cpumanager¶
- (2022) kubernetes.io: Kubernetes v1.26: CPUManager goes GA [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Celebrates the General Availability (GA) of the CPUManager in Kubernetes 1.26. This sub-system allows high-performance workloads to pin CPU cores, offering deterministic scheduling profiles by enforcing NUMA node alignment. Architects can exploit these improvements to drastically reduce latency in database pipelines, streaming engines, and high-frequency trading workloads.
Pod Resize¶
- (2023) kubernetes.io: Kubernetes 1.27: In-place Resource Resize for Kubernetes Pods (alpha) [MARKDOWN CONTENT] [ADVANCED LEVEL] [EMERGING] โ Reviews the highly anticipated alpha feature in Kubernetes 1.27: In-place Resource Resize for Pods. This allows engineers to dynamically scale CPU and memory resources allocated to running containers without requiring a pod restart. This capability drastically improves availability and operational costs for scaling-heavy stateful applications and microservices.
Scheduling (1)¶
Resource Allocation (2)¶
- (2022) kubernetes.io: Kubernetes 1.26: Pod Scheduling Readiness [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Explores the newly introduced Pod Scheduling Readiness feature in Kubernetes 1.26. This allows external controllers to declare scheduling gates on a pod, signaling the scheduler to ignore the resource until specific dependencies are resolved (e.g., custom network setups or resource pre-allocation). This significantly reduces scheduler queue overhead in dynamic, automated environments.
Kubernetes Ecosystem¶
Releases (2)¶
Etcd Updates¶
- (2026) Announcing etcd v3.7.0 [EN CONTENT] [ADVANCED LEVEL] ๐๐๐๐ [DATA-STORE] [EMERGING] [ENTERPRISE-STABLE] [ETCD] [RELEASE] โ The release of etcd v3.7.0 marks a significant milestone for the foundational key-value store underpinning Kubernetes cluster state. Grounded analysis of the release details indicates advancements in performance and operational stability for high-scale environments. From a curator's perspective, staying current with etcd versions is critical to maintain consensus reliability and leverage upstream security enhancements within modern platform engineering architectures.
Kubernetes Networking¶
Traffic Engineering¶
Load Balancing¶
- (2022) kubernetes.io: Kubernetes v1.26: Advancements in Kubernetes Traffic Engineering [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Highlights advancements in Kubernetes Traffic Engineering during the 1.26 release lifecycle, focusing on Service Internal Traffic Policy and Topology Aware Routing. These adjustments allow clusters to minimize latency and cross-AZ data transfer fees by dynamically routing traffic closer to the requesting endpoint. Excellent resource for networking engineers designing large-scale cloud topologies.
Kubernetes Operators And Controllers¶
- blog.runx.dev: Will That Kubernetes v1.22 Upgrade Break My Application? ๐๐๐๐๐ [DE FACTO STANDARD] โ Technical documentation and implementation reference for blog.runx.dev: Will That Kubernetes v1.22 Upgrade Break My Application? covering Kubernetes Operators and Controllers in Kubernetes Releases (blog.runx.dev).
- analyticsindiamag.com: Kubernetes v1.21 Released: Major Updates & Latest Features ๐ [COMMUNITY-TOOL] โ Technical documentation and implementation reference for analyticsindiamag.com: Kubernetes v1.21 Released: Major Updates & Latest Features covering Kubernetes Operators and Controllers in Kubernetes Releases (analyticsindiamag.com).
- inder-devops.medium.com: Itโs Time to Migrate your Container Runtime, Kubernetes 1.24 is coming ๐ [COMMUNITY-TOOL] โ Technical documentation and implementation reference for inder-devops.medium.com: Itโs Time to Migrate your Container Runtime, Kubernetes 1.24 is coming covering Kubernetes Operators and Controllers in Kubernetes Releases (inders-ai.medium.com).
- itnext.io: Unleashing the Power of Kubernetes 1.26: Exploring the New ValidatingAdmissionPolicy Feature with CEL ๐ [COMMUNITY-TOOL] โ Technical documentation and implementation reference for itnext.io: Unleashing the Power of Kubernetes 1.26: Exploring the New ValidatingAdmissionPolicy Feature with CEL covering Kubernetes Operators and Controllers in Kubernetes Releases (itnext.io).
- levelup.gitconnected.com: Kubernetes Planternetes v1.28: Non-Graceful Node Shutdown Feature ๐ [COMMUNITY-TOOL] โ Technical documentation and implementation reference for levelup.gitconnected.com: Kubernetes Planternetes v1.28: Non-Graceful Node Shutdown Feature covering Kubernetes Operators and Controllers in Kubernetes Releases (levelup.gitconnected.com).
Kubernetes Orchestration¶
Batch Jobs¶
Job Tracking¶
- (2022) kubernetes.io: Kubernetes 1.26: Job Tracking, to Support Massively Parallel Batch Workloads, Is Generally Available [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Details the GA milestone of Scalable Job Tracking in Kubernetes 1.26. By offloading job tracking from querying all remaining pods to an optimized finalizer architecture, the API server can efficiently orchestrate massively parallel processing workloads. This enhancement dramatically lowers memory consumption and API server load in large-scale machine learning and batch-processing frameworks.
Kubernetes Security¶
Policies¶
Deprecation Critique¶
- (2022) macchaffee.com: The Fumbled Deprecation of PodSecurityPolicies [ADVANCED LEVEL] [LEGACY] โ This architectural critique analyzes the friction caused by the rapid deprecation of PodSecurityPolicies (PSP). The author highlights gaps in transition paths and the operational challenges encountered by platform engineering teams moving legacy workloads to alternative admission controllers like Gatekeeper or Kyverno. It serves as an informative case study on upstream platform stewardship and API migration management.
Pod Security Policy¶
- (2022) kubernetes.io: PodSecurityPolicy: The Historical Context ๐ [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ A detailed post-mortem on PodSecurityPolicy (PSP) providing the architectural context behind its deprecation and eventual removal in v1.25. The publication details structural flaws in PSP's design, including complex API authorization models and the lack of a dry-run capability. This serves as critical foundational reading for security architects migrating clusters to modern Pod Security Standards (PSS) or third-party engines.
Kubernetes Storage¶
CSI¶
Volume Mounts¶
- (2022) kubernetes.io: Kubernetes 1.26: Support for Passing Pod fsGroup to CSI Drivers At Mount Time [MARKDOWN CONTENT] [ADVANCED LEVEL] [COMMUNITY-TOOL] โ Details the Kubernetes 1.26 enhancement enabling Pod fsGroup parameters to be passed directly to CSI drivers during volume mounting operations. By delegating directory permissions and ownership shifts to the underlying storage provider rather than executing local chown recursions, this update dramatically speeds up pod boot times on massive datasets. This optimization represents a major advancement in storage management efficiency.
Managed Kubernetes¶
AWS EKS¶
V1.22 (4)¶
- (2022) aws.amazon.com: Amazon EKS now supports Kubernetes 1.22 [COMMUNITY-TOOL] โ Curator Insight covers EKS supporting Kubernetes 1.22. Live grounding in 2026 treats this as historical information; modern EKS runs vastly newer versions, but this article highlights EKS migration paths and deprecation management.
EKS¶
Upgrades¶
- (2022) datree.io: EKS 1.22 Upgrade Tutorial [GUIDE] [LEGACY] โ An operations-oriented technical walkthrough focusing on upgrading Amazon EKS clusters to version 1.22. It documents the impact of deprecated v1beta1 API removals, specifically detailing transitions for Ingress and CustomResourceDefinition objects. SREs can leverage this guide to configure validation scripts and preventative gatekeeping within CI/CD pipelines prior to AWS control-plane updates.
๐ก Explore Related: Serverless | OCP 4 | Openshift
๐ See Also: Golang | Other Awesome Lists