Roadmap
LITMUS ROADMAP¶
This document captures only the high level roadmap items. For the detailed backlog, see issues list.
Completed¶
- Declarative Chaos Intent via custom resources
- Chaos Operator to orchestrate chaos experiments
- Off the shelf / ready chaos experiments for general Kubernetes chaos
- Self sufficient, Centralized Hub for chaos experiments
- Per-experiment minimal RBAC permissions definition
- Creation of 'scenarios' involving multiple faults via Argo-based Chaos Workflows (with examples for microservices apps like podtato-head and sock-shop)
- Cross-Cloud Control Plane (Litmus Portal) to perform chaos against remote clusters
- Helm3 charts for LitmusChaos (control plane and experiments)
- Support for admin mode (centralized chaos management) as well as namespaced mode (multi-tenant clusters)
- Continuous chaos via flexible schedules, with support to halt/resume or (manual/conditional) abort experiments
- Generation of observability data via Prometheus metrics and Kubernetes chaos events for experiments
- Steady-State hypothesis validation before, during and after chaos injection via different probe types
- Support for Docker, Containerd & CRI-O runtime
- Support for scheduling policies (nodeSelector, tolerations) and resource definitions for chaos pods
- Support for ARM64 nodes
- Scaffolding scripts (SDK) to help bootstrap a new chaos experiment in Go, Python, Ansible
- Support orchestration of non-native chaos libraries via the BYOC (Bring-Your-Own-Chaos) model
- Support for OpenShift platform
- Integration tests & e2e framework creation for control plane components and chaos experiments
- Documentation (usage guide for chaos operator, resources & developer guide for new experiment creation)
- Add architecture details & design resources
- Define community sync up cadence and structure
In-Progress (Under Active Development)¶
- Support for all ChaosEngine schema elements within workflow wizard
- Workflow YAML linter addition
- Minimized role permissions for Chaos Service Accounts
- Chaos-center users account to chaosService account map
- Provide complete workflow termination/abort capability
- Cross-hub experiment support within a Chaos Workflow
- Helm Chart for Chaos Execution Plane
- Enhanced CRD schema for ChaosEngine to support advanced CommandProbe configuration
- Support for S3 artifact sink (helps performance/benchmark runs)
- ChaosHub refactor for 2.x user flow
- Chaos experiments against virtual machines and cloud infrastructure (AWS, GCP, Azure, VMWare, Baremetal)
- Improved documentation and tutorials for Litmus Portal based execution flow
- Off the shelf chaos-integrated monitoring dashboards for application chaos categories
- Support for user defined chaos experiment result definition
- Increased fault injection types (IOChaos, HTTPChaos, JVMChaos)
- Special Interest Groups (SIGs) around specific areas in the project to take the roadmap forward
Backlog¶
- Pre-defined chaos workflows to inject chaos during application benchmark runs
- Support for cloudevents compliant chaos events
- Improved application Chaos Suites for various CNCF projects