sajjadm624/Kubecrash

# KubeCrash     ## What is KubeCrash? KubeCrash is a **Kubernetes incident training platform** designed to turn learners into confident operators. It combines two modes in one experience: - **Incident Game** — pressure-tested troubleshooting with real `kubectl` command flow - **CKA Learning Journey** — structured domain-by-domain progression across the CKA blueprint No local cluster setup required. Everything runs in the browser, with simulation logic tuned for practical decision-making. ## Quick start Run KubeCrash locally in minutes. ### 1. Backend (FastAPI + WebSocket) python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate pip install -r backend/requirements.txt uvicorn backend.main:app --reload --port 8000 ### 2. Frontend (React + Vite) cd frontend npm install npm run dev Open **http://localhost:5173** ## Incident Game — Command loop 1. Pick an incident from the level select screen 2. Read the briefing like an on-call handoff 3. Run `kubectl` commands to isolate root cause 4. Apply the fix before the timer expires 5. Submit and compare results on the leaderboard ### Levels | # | Title | Concept | |---|-------|---------| | 1 | The Crash at Dawn | CrashLoopBackOff, env vars | | 2 | The Invisible Service | Label selectors, endpoints | | 3 | The OOM Reaper | OOMKilled, resource limits | | 4 | The Ghost Image | ImagePullBackOff, rollback | | 5 | The Dead Node | Node lifecycle, drain | ## CKA Learning Journey Structured CKA preparation built for operational fluency, not passive reading. ### Features - 15 structured lessons across **Beginner → Foundation → Intermediate** tracks - Interactive shell with **simulated `kubectl` output** for each command path - Per-checkpoint **why-this-command** explanations for fast mental model building - **Command syntax coach** with live verb/resource/flags breakdown - **Lesson recap quiz** to lock in understanding before moving forward - Timed **mini-mocks** and a full **120-minute weighted CKA mock** - **Adaptive hint modes**: `beginner`, `standard`, `exam`, `adaptive` - **Realtime architecture diagram** that updates per lesson domain - Persistent score, streak, badges, and certificate state via localStorage - Official Kubernetes docs linked directly from lessons - Built-in 30-day study roadmap ### CKA blueprint coverage | Domain | Weight | |--------|--------| | Troubleshooting | 30% | | Cluster Architecture, Installation and Configuration | 25% | | Services and Networking | 20% | | Workloads and Scheduling | 15% | | Storage | 10% | ### Lesson tracks | Track | Lessons | |-------|---------| | Beginner | Lesson 0: Kubernetes from Zero | | Foundation | Lessons 1–6: Env vars, Services, Resources, RBAC, PVCs, Ingress | | Intermediate | Lessons 7–14: Taints, Rollouts, ConfigMaps, StatefulSets, DNS, Upgrades, TLS | ## KubeCrash Mastery Roadmap KubeCrash began with focused incident scenarios to make onboarding approachable. KubeCrash now expands into a full mastery platform with measurable skill growth, portfolio evidence, and role-based pathways. ### Product evolution map | Stage | Experience | Scope | |------|------------|-------| | 1. Onboarding | Incident game intro | 5 fast starter incidents | | 2. Core Training | CKA learning journey | 15 lessons + mocks | | 3. Advanced Tracks | Incident case-study academy | 4 tracks x 4 lessons | | 4. Mastery Platform | Role paths + capstones + skill graph | 60+ labs + 5 projects | ### Curriculum target | Layer | Target count | Outcome | |------|--------------|---------| | Starter incidents | 10 | Build initial confidence in command fluency | | Foundation labs | 30 | Strong CKA fundamentals across all blueprint domains | | Advanced incidents | 24 | Multi-signal diagnosis under realistic constraints | | Role-path missions | 16 | SRE, Platform, Security, DevOps specialization | | Capstone projects | 5 | Portfolio-grade end-to-end Kubernetes projects | ### Role paths (new) - SRE Path: observability, SLOs, alerting, incident command, postmortems - Platform Engineer Path: cluster operations, cost controls, multi-tenant architecture - Security Engineer Path: RBAC, policy, secrets, supply-chain and audit controls - DevOps/GitOps Path: release strategy, progressive delivery, rollback governance ### Mastery model and progression rules Learners unlock new content by proof of capability, not just completion. #### Skill graph nodes - Command fluency - Debug workflow - Workload reliability - Networking diagnostics - Storage reliability - Security hardening - Observability reasoning - Delivery safety - Cluster operations #### Unlock logic (default) 1. Foundation Track unlocks immediately 2. Advanced Track requires: - 70%+ completion in Foundation - Minimum 60% average quiz score in completed lessons 3. Role Path requires: - One completed Advanced Track - At least 3 saved retrospectives 4. Capstone requires: - Two completed Role Paths - Mastery score >= 75 in at least 5 skill nodes ### Scoring model Total score combines speed, correctness, and learning behavior. $$ Mastery = 0.35C + 0.20S + 0.20R + 0.15Q + 0.10L $$ Where: - $C$ = command correctness score - $S$ = scenario completion reliability score - $R$ = retrospective quality/completion score - $Q$ = quiz understanding score - $L$ = long-term retention score (repeat challenge delta) ### Portfolio outputs (must-have) Every advanced lesson and capstone should produce artifacts: - Incident brief + timeline - Root cause analysis - Retrospective answers - Action items - Suggested runbook snippets - Final score + elapsed time ### 12-week release plan | Week | Delivery focus | Exit criteria | |------|----------------|---------------| | 1 | Expand starter incidents from 5 to 8 | New incidents playable end-to-end | | 2 | Add 2 more starter incidents (10 total) | Starter onboarding complete | | 3 | Add 8 foundation labs (phase A) | Lessons + checks + recap quizzes live | | 4 | Add 8 foundation labs (phase B) | 16 new labs total in roadmap branch | | 5 | Add 8 foundation labs (phase C) | 24 new labs total | | 6 | Add 6 foundation labs + polish | 30 foundation labs complete | | 7 | Build role path framework + progress rules | Path UI + unlock gating functional | | 8 | Ship SRE + Platform paths | 8 role missions live | | 9 | Ship Security + DevOps paths | 16 role missions complete | | 10 | Implement skill graph + mastery scoring | Node scores visible in profile | | 11 | Build first 3 capstones | Project rubrics + artifact export | | 12 | Build final 2 capstones + launch prep | 5 capstones + launch readiness | ### Success metrics - 7-day retention > 35% - Lesson-to-lesson completion > 60% - Advanced track completion > 30% of active learners - Average mastery score gain of +20 points in 30 days - At least 1 portfolio artifact exported per active learner each week ### Definition of done for KubeCrash - 60+ labs live with stable validation - 4 role paths with progression gates - 5 capstone projects fully rubric-scored - Skill graph and mastery score visible in UI - Exportable learner portfolio artifacts ### Revamp execution docs - [Execution plan](docs/revamp/EXECUTION_PLAN.md) - [Sprint backlog](docs/revamp/SPRINT_BACKLOG.md) - [Lab spec template](docs/revamp/LAB_SPEC_TEMPLATE.md) - [Capstone rubric](docs/revamp/CAPSTONE_RUBRIC.md) ## Tech stack | Layer | Technology | |-------|-----------| | Frontend | React 18, Vite, Zustand, xterm.js | | Backend | FastAPI, Uvicorn, WebSockets, Pydantic | | Simulation | Custom kubectl parser + per-level state machines | | Persistence | localStorage (client-side progress) | ## Project structure KubeCrash/ ├── backend/ │ ├── main.py # FastAPI app + WebSocket handler │ ├── engine/ # Kubectl parser + scenario engine │ ├── scenarios/ # Per-level incident definitions │ └── routers/ # HTTP endpoints (leaderboard, session) └── frontend/ └── src/ ├── components/ # LearningJourney, Terminal, LevelSelect ├── hooks/ # useTerminal (xterm lifecycle) ├── store/ # Zustand game state └── utils/ # kubectlParser (semantic matching) ## License MIT

标签：自定义脚本