shardingsphere-on-cloud icon indicating copy to clipboard operation
shardingsphere-on-cloud copied to clipboard

Published 20 hours ago • verified publisher icon apache

Introduce ChaosEngineering for ShardingSphere

Open mlycore opened this issue 2 years ago • 4 comments

As a critical components for any production use, ShardingSphere plays a very important role which have a high level availability requirements.

mlycore avatar Aug 22 '22 04:08 mlycore

At first we plan to have the DatabaseChaos, providing some tests on high availability of ShardingSphere. DatabaseChaos is currently a CRD of DatabaseMesh supporting AWS RDS Instance rebooting and AWS RDS Cluster automatically failover.

mlycore avatar Dec 30 '22 01:12 mlycore

Currently we plan to support two types of chaos for ShardingSphere:

  • Infra: like Kubernetes and AWS chaos
  • JVM: general JVM chaos

mlycore avatar Feb 22 '23 06:02 mlycore

Here are the issues:

  • #253
  • #254

mlycore avatar Mar 09 '23 05:03 mlycore

Recent thoughts on this, propose a new CRD called ShardingSphereChaos that contains two different kinds of Chaos:

  • General Chaos: including kubernetes, network, disk, time, etc.
  • JVMChaos: including byteman/bytebuddy scripts.

We have to specify the detail of one Chaos, the execution requirement, the explosive radius, the load generator, the observed result and its regarding quality gate. Then the chaos would be a full lifecycle management.

mlycore avatar Mar 16 '23 05:03 mlycore