WFGY/ProblemMap/GlobalFixMap/OpsDeploy/read_only_mode_and_maintenance_window.md

Read-Only Mode and Maintenance Window — OpsDeploy

🧭 Quick Return to Map

You are in a sub-page of OpsDeploy.
To reorient, go back here:

• OpsDeploy — operations automation and deployment pipelines
• WFGY Global Fix Map — main Emergency Room, 300+ structured fixes
• WFGY Problem Map 1.0 — 16 reproducible failure modes

Think of this page as a desk within a ward.
If you need the full triage and all prescriptions, return to the Emergency Room lobby.

Freeze writes quickly and communicate clearly during incidents, migrations, or index rebuilds. Keep retrieval online while protecting data integrity and user trust.

Open these first

• Ordering and cold boots: Bootstrap Ordering, Pre-Deploy Collapse
• Idempotency and dedupe: Idempotency & Dedupe
• Incident comms: Incident Comms & Statuspage
• Cache and warmup: Cache Warmup & Invalidation

Acceptance targets

• All write endpoints return 423 or 503 with Retry-After
• No production writes in storage audit during the window
• Read paths healthy: error budget unaffected
• Banner or Statuspage posted before and after with exact times
• ΔS and λ trends flat during the window for read requests

60-second checklist

1. Flip read-only flag
   Global feature flag read_only=true. Middleware blocks writes, tool calls that mutate state, and ingestion jobs.
2. Communicate
   Post Statuspage and in-product banner with start and end times. Include contact and rollback plan.
3. Queue writes
   Buffer allowed write intents to a durable queue with idempotency keys. Drop non-critical writes.
4. Headers
   Write endpoints respond with Retry-After: <seconds> and a stable error code for clients to respect.
5. Drains
   Stop background writers, disable compaction and long migrations during the window.

Minimal playbook

• Scope: enumerate write surfaces including embeddings backfills, vector upserts, tool call side effects, KV mutations.
• Guard: enforce in API gateway and service layer. Return machine-readable errors.
• Observe: track blocked write count, queue depth, read latency, ΔS and λ.
• End: re-enable writes, run a short soak, publish closure notice, then backfill the queued intents.

Common pitfalls → fix

• A background job keeps writing
  → include job scheduler in the flag check and stop crons.
• Clients spin on retries
  → add jittered backoff and a max-retry budget. See Retry Backoff.
• Cached stale banners
  → set short TTL and purge caches on start and end.

Escalate

If read errors or ΔS spikes appear, extend the window and execute a limited Rollback & Fast Recovery.

---

🔗 Quick-Start Downloads (60 sec)

Tool	Link	3-Step Setup
WFGY 1.0 PDF	Engine Paper	1️⃣ Download · 2️⃣ Upload to your LLM · 3️⃣ Ask “Answer using WFGY + <your question>”
TXT OS (plain-text OS)	TXTOS.txt	1️⃣ Download · 2️⃣ Paste into any LLM chat · 3️⃣ Type “hello world” — OS boots instantly

---

🧭 Explore More

Module	Description	Link
WFGY Core	WFGY 2.0 engine is live: full symbolic reasoning architecture and math stack	View →
Problem Map 1.0	Initial 16-mode diagnostic and symbolic fix framework	View →
Problem Map 2.0	RAG-focused failure tree, modular fixes, and pipelines	View →
Semantic Clinic Index	Expanded failure catalog: prompt injection, memory bugs, logic drift	View →
Semantic Blueprint	Layer-based symbolic reasoning & semantic modulations	View →
Benchmark vs GPT-5	Stress test GPT-5 with full WFGY reasoning suite	View →
🧙‍♂️ Starter Village 🏡	New here? Lost in symbols? Click here and let the wizard guide you through	Start →

---

👑 Early Stargazers: See the Hall of Fame —
⭐ WFGY Engine 2.0 is already unlocked. ⭐ Star the repo to help others discover it and unlock more on the Unlock Board.