[translation] chaos mesh moved to CNCF incubator

Published on 2022 year 2 month 16 Japan

CNCF Technical Supervision Committee （TOC） Have voted to accept Chaos Mesh As CNCF The incubation program of .

Chaos Mesh Originally as an open source distributed database TiDB Test platform of , It is a multifunctional chaos engineering platform , Can be in Kubernetes Coordinated chaos experiment in environment . It helps identify potential fault points , Help ensure Kubernetes Infrastructure can withstand accidental damage .

since 2020 year 7 Month is accepted into CNCF Since sandbox ,Chaos Mesh Two major versions have been implemented （v1.0 and v2.0） and 30 A minor version , In observability 、 Great improvements have been made in function and security . Some prominent features include .

• Chaos dashboard , A visual aid , Help users manage and monitor chaotic experiments through the network user interface .
• Native workflow engine , Used to define chaotic scenes , Manage the state check of chaos experiment group and Application .
• More powerful and comprehensive chaos simulation , cover StressChaos、DNSChaos、JVMChaos、AWSChaos、GCPChaos、HTTPChaos etc. .
• be based on Kubernetes RBAC Authorization mechanism of permission policy .

exceed 50 Organizations have adopted Chaos Mesh To test and improve the flexibility of its distributed system . Adopters include ByteDance、DataStax、Percona、Prudential、NetEase Fuxi、RabbitMQ、SHAREit、XPeng Motors etc. . Besides , Microsoft Azure Chaos Studio When the cloud provider has take Chaos Mesh Integrate To their SaaS In solution , Allow their users to inject faults AKS In the cluster . Many of these companies Also giving feedback for the project .

"Chaos Mesh Based on the Kubernetes CRD On , This has given the project a head start ," Netease Fuxi Zhang Hui, a senior quality assurance engineer, said ." It provides fine-grained chaos support , A unified UI-- Chaos dashboard , And enhanced observability and accurate chaos range control . All this is done by an open 、 Collaborative and dynamic community driven ."

"Chaos Mesh It provides rich fault simulation methods , Help the infrastructure team verify high availability in advance 、 Scenarios such as network traffic loss and bidirectional synchronization , This helps to mine the defects of the solution , Reduce risk ,"XPeng Motors Chief engineer of Hengliang Tan say ." It can also help our team reduce testing costs ".

" We set up Chaos Mesh My mission is simple -- Make chaos engineering easier , Make complex systems resilient , Because they should be like this ,"Chaos Mesh Defenders and co creators of Cwen Yin say ." The power of communities and ecosystems is critical to achieving this goal . We are glad to see Chaos Mesh Become an incubation project .CNCF It is the driving force of cloud native ecosystem , With the support and guidance of the community , Will help us further develop Chaos engineering ."

Chaos Mesh use Kubernetes framework , Use Kubernetes CustomResourceDefinitions（CRD） To define chaotic objects . It can also work with other cloud native projects （ Such as Argo、Grafana and Prometheus） Tight integration , Make chaos experience more manageable 、 Customizable and observable .

Main components .

• Chaos Dashboard.Chaos Mesh The visualization component of .Chaos Dashboard It provides a set of user-friendly network interfaces , Users can operate and observe through it Chaos experiment .
• Chaos Controller Manager.Chaos Mesh The core logic component of .Chaos Controller Manager Mainly responsible for scheduling and management Chaos experiment .
• Chaos Daemon. The main execution components .Chaos Daemon stay DaemonSet Run in mode , Default ownership Privileged jurisdiction （ You can disable ）.
• Chaosd： One for non Kubernetes Toolkit for node injection failure .

Significant milestones .

• 4.5K GitHub The star
• 1.3K Submission quantity
• 800 Multiple closed issues
• come from 60 Multiple organizations 125 Contributors
• 32 Secondary release
• 50 Multiple adopters

"CNCF Chief technology officer Chris Aniszczyk say ：" No cloud native deployment is perfect -- Failure always happens , Therefore, using chaos engineering to establish a flexible culture can save time and money for the organization ." We are looking forward to seeing how chaos network grows as an incubation project , And have an impact on the situation in the field of chaos and elastic Engineering ."

Chaos Mesh There is a complete The roadmap , The team is actively adding new features and functions , At the same time, improve the whole chaotic experience . The team is working hard to provide an efficient status verification mechanism and reporting capability , To improve usability and observability . Continuous optimization of the workflow engine will enable users to pass Chaos Mesh Achieve a complete chaotic engineering cycle . Rich supported fault types , To cover as many actual failures in the cloud native system as possible , Will improve functionality and scalability . Besides , The team will provide a plug-in mechanism , Allow users to freely expand their own fault types , And publish their own plug-ins and chaotic scenes to Chaos Mesh On .
As a CNCF Managed projects ,Chaos Mesh It is part of a neutral foundation consistent with its technical interests , It's also bigger Linux Part of the foundation , The latter provides management 、 Marketing support and community promotion .Chaos Mesh Added incubation Technology Argo, Buildpacks, CloudEvents, CNI, Contour, Cortex, CRI-O, Dragonfly, emissary-ingress, Falco, Flux, gRPC, KEDA, KubeEdge, NATS, Notary, OpenTelemetry, Operator Framework, Rook, SPIFFE, SPIRE, and Thanos, Maturity requirements for each level , Please visit CNCF Graduation criteria .