Revolutionizing CICD with Kubernetes.
It'd be sweet to have a CI/CD platform that doesn't require constant jerry-rigging to work the way you need it to. Your automation tools should just work,
Enter Cruiser, a robust approach to CI/CD, with enterprise-grade security and extensibility. With native Kubernetes support and an interface made for humans, it's the next best thing in DevSecOps!
The project is currently in its early alpha stages. We are in the process of solidifying the base of the framework and stabilizing the architecture. There will be bugs, and you have been warned :)
Ordered by priority.
- Add CRON triggers
- Add "nightly" additional that's fixed on system time
- Pipeline instance labels can be defined from simple JS scripts
- Support Pipeline definition as code file (YAML, JSON)
- Pipeline, stage, job failure hooks
- Projects that can group pipelines and teams / boards
- Env vars & secrets on "global system"
- Env vars & secrets on "project"
- Auto-trigger and approve next release if previous release has been deployed and not redacted
- Detect OOM crashes from known languages / build tools
- Add support for pipeline templates
- Add support for non-git based code repos
- Create base worker images and sample k8s setup
- Record error messages over time and provide "smart" analysis
- Known build tools (docker, gulp, msbuild etc) should have a tailored experience
- Support for setting "Deployment freeze" periods
- Support for "rollback to latest stable build" API
- Support for watching deployment progress
- Support for plugin system
- Support for hosting git repos
- Support user metrics for hosted repos
- Support Artillery load testing + metrics (builtin?)
- Support for common load testing & quality gate metric integration
- Documentation -- need help!
- Agent configure graph: disabled tasks become opaque
- Agent configure graph: task conditions (if not met make the dependants all a shade of red/pink?)
- On startup run preflight system checks
- System alerts -- failed pipelines, system state + health
- Allow rollback of failed / partial deployments / tasks
- Simulate task execution (clarify: how?)
- Branch based templating and filters
- Advanced task filter syntax (use JS?)
- Kanban tracking board
- metrics export => how long do builds take, how long does each stage/job/task take
- metrics export => how many builds were manual, automated, scheduled
- Fetch secrets from a key vault
- Plugins (ui, agent, backend)
- AI auto build stages and pipelines
- Drop in jenkins replacement or conversion tool
- Create Kube CRDs that define everything about the server + build agents
- Create Kube CRDs that define build and release pipelines
- Static build agent support
- Destroy all humans
- Add stage "Triggers"
- Add stage "Approvers"
- Add stage webhook support
- Env vars & secrets on "pipeline"
- Env vars & secrets on "stage"
- Env vars & secrets on "job"
- Env vars & secrets on "task group"
- Env vars & secrets on "task"
- Enable disabling pipelines, jobs, stages, task groups etc.
- Create breakpoints and allow remote terminal connections through web UI