Welcome to the Oracle Cloud Infrastructure (OCI) Container Services newsletter. Here we will share product updates, technical guidance, and practical examples to help you build, run, and scale containerized workloads on OCI. We’re glad you’re here, and we hope this newsletter helps you get more from OCI Kubernetes Engine (OKE), OCI Container Instances, OCI Functions, and the broader OCI container ecosystem.

OCI Kubernetes Engine    Container Instances    OCI Functions  

Karpenter Provider for OKE goes GA 

The Karpenter Provider for OCI has gone GA. This release makes Kubernetes autoscaling on OCI simpler, more flexible, and more efficient. Instead of scaling a predefined node pool, Karpenter watches for unscheduled pods and provisions the right compute shape in real time. Platform teams will appreciate how they can spend less time planning node pools up front and managing node sprawl later. Application teams will like that they can keep using normal Kubernetes requests, selectors, taints, and affinities without waiting on the platform team to create a new pool every time a workload changes.

Learn more: Smarter Kubernetes autoscaling on OCI with Karpenter provider now GA

Disaster Recovery and Containers 

Customers rely on disaster recovery to keep their business running when an outage disrupts operations. In this set of OCI articles, Oracle shares practical cross-region recovery patterns for OCI Functions, OCI Container Instances, and OCI Kubernetes Engine (OKE), covering everything from recreating serverless workloads in a destination region to replicating container images, synchronizing state, preparing standby environments, and validating failover readiness. Together, the articles show how teams can move beyond static runbooks and build a tested recovery posture that is aligned to real-world service disruptions.

Learn more: 

OKE Welcomes Kubernetes 1.35: What’s new, what’s changing, and what to do next

Oracle Kubernetes Engine now supports Kubernetes 1.35, bringing new capabilities for day-2 operations, security, scheduling, and networking, along with important upgrade considerations for existing clusters. [Placeholder: add one sentence here highlighting OKE-specific news related to Kubernetes 1.35.]

Kubernetes 1.35 also adds to the urgency of moving off Oracle Linux 7 (OL7). OCI Blogs have already covered key milestones for OL7 on OKE: support for OL7 on ARM ended on January 1, 2025, and OL7 x86_64 entered Extended Support through June 2028. Kubernetes 1.35 adds fresh pressure because it requires adoption of cgroups v2, which OL7 does not support. Now is the time to start moving to Oracle Linux 8.

Learn more:

Nginx controller retirement

OKE has updated its NGINX Ingress Controller documentation to note that the Kubernetes community plans to stop maintaining the NGINX Ingress Controller after March 2026. Existing deployments are expected to keep working, but without future bug fixes or security updates, teams should start thinking about their next ingress path: either another Ingress controller or a move to the Kubernetes Gateway API.

Learn more: Example: Setting Up an Nginx Ingress Controller on a Cluster

Crossplane Provider for OCI Update 

OCI has significantly expanded the Crossplane Provider for OCI for teams adopting Kubernetes-native infrastructure management. Recent updates include a modular provider family architecture, support for Instance Principal and Workload Identity, coverage for 150+ OCI services, and pre-built images that simplify deployment. These improvements make it easier for platform teams to provision and manage OCI resources through Kubernetes APIs while strengthening security, modularity, and operational consistency.

The full post also looks at Crossplane’s growing momentum, explains what these updates mean for OCI users, and walks through how to get started. Read the article for a quick deployment guide, an Object Storage example, and a preview of upcoming capabilities aligned with the Crossplane roadmap.

Learn more: 

Where OCI Functions Is Showing Up in SaaS Delivery

OCI Functions adoption is growing across SaaS and services-led projects. These customers are using OCI Functions as a lightweight layer for authorization, document processing, event routing, payload transformation, and secure data handling between systems. OCI Functions works for them because it enables custom logic to connect and extend workflows, without standing up a full service for every step. Come read a collection of practical design lessons around keeping functions narrowly scoped, using Object Storage for document-heavy flows, and planning subnet capacity early for private deployments. We’re turning those recurring patterns into clearer guidance so more teams can adopt them with less trial and error.

Learn more: OCI Functions in Customer Workflows: Common Patterns and Practical Design Checklists

Technical Documentation Updates

OCI Documentation continues to be a critical resource for learning. In the past quarter, we’ve added documentation for: 

  • Kubernetes 1.35.0, 1.34.2, and 1.32.10 support here
  • the AMD GPU Plugin cluster add-on here
  • dynamically provisioning a PVC on a new Lustre file system here
  • when provisioning load balancers and network load balancers for services of type LoadBalancer:
    • how to enable instant failover for network load balancers here
    • how to use the OCI Certificates service to manage certificates for frontend SSL here
    • how to specify the compartment in which load balancers and network load balancers are created here
    • how to specify the maximum number of allowed load balancer connections to backend servers here