# clar200-unlock

> Unlock EMC CLAR200-locked Exos X18 SAS drives via the Seagate F3 terminal, using an ESP32-C3 bridge.

## Start Here

**Read the latest handoff first:** `.claude/handoffs/` (most recent file).
Then `PROCEDURE.md` (the runbook), `IDEA.md` (brief), `DECISIONS.md` (settled choices).
Background on why host-side tools fail: `~/bin/EMC_CLAR200_DRIVE_FIX_NOTES.md`.

## Project Identity

Two Seagate Exos X18 SAS drives (ST16000NM004J, EMC fw ES2264 — SN ZRS015V7,
ZR511KWK) are locked in the "SS160520 CLAR200" state: SCSI INQUIRY/VPD work, but
every other command (incl. READ CAPACITY) returns Hardware Error. No host tool
can format them. This project drives the drives' out-of-band **F3 service UART**
(via an ESP32-C3 as a 3.3 V USB-serial bridge) to reformat them to 512-byte
sectors and reclaim them as 14.6 TB+ spares for the `tank` pool.

## Current State

- **Phase:** building → first hardware attempt pending
- **Repo:** https://git.sethpc.xyz/Seth/clar200-unlock
- **Hardware:** ESP32-C3 on steel141 `/dev/ttyACM0` (MAC 58:8c:81:ac:f0:4c).
  Drive ZRS015V7 in a pve173 JBOD (must be benched for F3 work).
- **Done:** firmware (`firmware/clar200_bridge.ino`) + flash script + full
  procedure written.
- **Next:** `./firmware/flash.sh`, sanity-test bridge, bench drive, find F3
  pinout, get `F3 T>`, format to 512.

## Conventions

- **Safety:** bench-only F3 work — NEVER in the pve173 JBOD next to live tank.
  3.3 V only; share TX/RX/GND, never power. `screen -L` to log every session.
  Escalation brake applies: silent terminal → re-check wiring/baud, never a more
  aggressive format command.
- **F3 honesty:** Exos X18 dialect ≠ 7200.11. Map commands with `?` before
  firing; record discovered tokens in `docs/f3-command-reference.md`.
- Conventional commits; commit + `gitea push` each meaningful change.