Tier B — knowledge transfer (demo script + environment checklist)
Primary docs: TIER_B_HANDOVER.md, E2E_AUTOMATION_PLAN.md, TIER_B_MAINTENANCE_CHECKLIST.md
1. Session goal (45–60 min)
Enable maintainers to:
- Understand the Tier B CI architecture (mocks + harness).
- Debug common failures using artifacts and the troubleshooting matrix.
- Operate monitoring and respond to the consecutive-failure alert.
- Know what Tier C adds and what remains out-of-scope for Tier B CI.
2. Demo script (suggested flow)
Part A — Architecture overview (5–10 min)
- Open
docs/e2e/TIER_B_HANDOVER.md and walk through:
- Components + ports table
- Mermaid sequence diagram (Backend ↔ SyntheticDesktop ↔ MockHelixApi)
- Troubleshooting matrix (where to look first)
Part B — Show the workflow and the Tier B boundary (10–15 min)
- Open
.github/workflows/e2e-test.yml.
- Call out:
- Tier A gate:
send_e2e_eventsub.py → GET /api/pool/me
- Tier B step:
run_e2e_tier_b.py
- Job summary: “E2E performance (this run)” table (total wall time + Tier B boundary time)
- Always-on artifact upload:
e2e-service-logs
Part C — Show MockHelixApi + SyntheticDesktop interaction (10–15 min)
- Explain the critical assertion points:
GET http://127.0.0.1:9053/last-request proves Helix send was attempted and captured
GET http://127.0.0.1:9054/last-run proves Desktop-equivalent choreography succeeded
- Tie these back to:
- Backend config
Twitch:HelixApiBaseUrl (root URL, not .../helix)
- Desktop API key header
X-MGM-ApiKey
Part D — Log artifacts and post-mortem workflow (10 min)
- From a failing run (or a saved artifact), walk through:
mgm-backend.log (startup and API errors)
mgm-tier-b-orchestrator.log (Python orchestration and assertions)
mgm-mock-helix.log (POST capture / auth errors)
mgm-synthetic-desktop.log (step-by-step status codes)
Part E — Monitoring + alert response (5–10 min)
- Weekly health report:
.github/workflows/e2e-weekly-health-report.yml → Summary table (rolling 30‑day stats)
- Consecutive failure alert:
.github/workflows/e2e-consecutive-failure-alert.yml → opens exactly one issue
- Response: download artifacts from the two linked failed runs, triage, document root cause, close issue when green again
3. Demo environment checklist
Minimal requirements (docs-only walkthrough)
- Access to GitHub repo Actions page
- A recent successful
e2e-test.yml run to open (for Summary + artifact shape)
- A past failing run (optional) or an archived artifact for log walkthrough
Local runnable environment (optional)
If you want to run the Tier B flow locally (mirrors CI):
- Software
- .NET SDK (per repo standard)
- Python 3 + pip
- PostgreSQL 16 (Docker or local install)
- Services and ports
- Backend:
http://127.0.0.1:8080
- MockEventSubWebhook:
:9051
- MockExtensionJwt:
:9052
- MockHelixApi:
:9053
- SyntheticDesktop:
:9054
- Config
ASPNETCORE_ENVIRONMENT=Development
Mgm:EnableE2eHarness=true
Mgm:ApiKey set consistently for Backend + SyntheticDesktop + orchestrator
Twitch:HelixApiBaseUrl=http://127.0.0.1:9053
- Dummy
Twitch:BroadcasterAccessToken, Twitch:BroadcasterUserId, Twitch:HelixClientId (non-empty)
- Verification scripts
pip install -r .github/scripts/tier_b_verification/requirements.txt
- Run
python3 .github/scripts/tier_b_verification/check_workflow_integration.py
- Run
python3 .github/scripts/run_e2e_tier_b.py --api-key <key>
4. Scheduling / invite template (fill in)
- Title: Tier B E2E — maintainer knowledge transfer
- Duration: 60 min
- Presenter: Anatoly Ivanov (
ai.vibeqodez@vk.com)
- Audience: CI maintainers, Backend maintainers, future Tier C implementers
- Proposed time: TBD
- Meeting link: TBD
- Preread:
docs/e2e/TIER_B_HANDOVER.md
docs/e2e/TIER_B_MAINTENANCE_CHECKLIST.md
docs/e2e/TIER_C_KICKOFF_PLAN.md
Live demo prep (if doing a live Actions walkthrough)
- Pick a known green
e2e-test.yml run URL ahead of time.
- Pick a known red run URL (or download and save
e2e-service-logs) for artifact walkthrough.
- Have the port matrix handy: 8080 / 9051–9054.