stackconf 2024 | Ignite: Is rust good for Kubernetes by Natalie Serebryakova .pdf
0 likes52 views
The document discusses the use of Rust for developing Kubernetes operators, highlighting its performance, memory management, and safety benefits compared to Go. It outlines the steps for building and deploying a Kubernetes operator for persistent volume claims (PVCs) and persistent volumes (PVs), including setup requirements and best practices. The summary concludes that the developed operator effectively ensures data persistence in Kubernetes by managing the storage infrastructure.
stackconf 2024 | Ignite: Is rust good for Kubernetes by Natalie Serebryakova .pdf
- 1. Is Rust good for Kubernetes? Natalie Serebryakova
- 2. Whoami ● My name is Natalie ● I'm a Staff Cloud Engineer ● LinkedIn: https://www.linkedin.com/in/login-id/ ● AWS Community Builder ● AWS User Group co-organizer
- 3. Agenda ● Kubernetes Operator ● Using Rust to develop Kubernetes Operator (PVC, PV) ● Demo using PVC Operator example ● Summary
- 4. Kubernetes Operators (Custom controllers) Operators let you use custom app-specific object types to automate the underlying Kubernetes configuration.
- 5. How Kubernetes Operators used? ● Kubernetes Operators are processes that connect to the Kubernetes API and watch for events (typically on a limited number of resource types). ● When a relevant event occurs, the operator reacts and performs a specific action
- 6. Why to use Rust for Kubernetes ecosystem? Feature Go Rust Performance ✅ Good enough ✅ low-level control over memory management ✅ compiler optimizations ✅ and concurrency features Concurrency ✅ Goroutines and channels for concurrency Async/await syntax : ✅ require finer control over concurrency ✅ error handling ✅ have performance-critical asynchronous tasks Safety ✅ Strong typing, but lacks memory safety checks ✅ Strong typing with rigorous (guaranteed) memory safety checks Majority K8s ecosystem (as of today) is build in Go. Is there benefit of using Rust?
- 7. Building Kubernetes Operators PVC with Rust • Kubernetes Operator can automate the management of PVCs and PVs, handling lifecycle events like creation, deletion, scaling, and backup/restoration To build Kubernetes Operators using Rust in Kubernetes clusters we will need to Extend Kubernetes API with Custom Resource Definitions (CRDs) to support • PersistentVolumeClaims (PVCs): PVCs request specific size and access modes (e.g., read/write) for storage from Persistent Volumes (PVs). • PersistentVolumes (PVs): PVs are the storage resources in the cluster. Each PV corresponds to a PVC's request.
- 8. Building and deploying K8s Operator with Rust ● Rust Environment: Ensure Rust and Cargo are installed. ● Kubernetes Cluster: A Minikube or any Kubernetes cluster. ● Docker: For building and pushing the operator's container image. ● kubectl: For interacting with the Kubernetes cluster. ● kube-rs, kuberntes-openapi, tokio, serde, log Crates: Required Rust crates.
- 9. Rust crates to use kube-rs and kubernetes-openapi are Rust crates specifically designed to support the development of Kubernetes operators. - kube-rs: ● Provides useful abstractions over Kubernetes API ● Simplifies handling of Custom Resource Definitions (CRDs) - kubernetes-openapi: ● Used by kube-rs for Kubernetes API interaction ● Enhancing operator development capabilities.
- 10. Best practices for writing Kubernetes Operators 1. Define clear Custom Resources (CRDs) and associated control loops 2. Follow K8s API conventions, client libraries 3. Handle state properly 4. Incorporate error handling 5. Secure your Operator 6. Document your Operator
- 11. Kubernetes Client configuration ● Initialize the Kubernetes client for interacting with the cluster, leveraging the default
- 12. Controller Components ● Struct holds configuration that affects the controller's behavior, such as default values for storage classes or thresholds for certain operations.
- 15. Packaging the operator “Package” into the container by creating a Dockerfile: Run : docker build -t yourusername/k8s_pvc_operator:v1 . docker push yourusername/k8s_pvc_operator:v1
- 16. Deploying the operator 1. kubectl apply -f local-pv.yaml 2. kubectl apply -f local-pvc.yaml
- 18. Summary Developed Operator ensures data persistence in a Kubernetes cluster by managing the underlying storage infrastructure .
- 19. THANK YOU