feat: add sshfs mount watchdog timer for auto-recovery

When zzz reboots or SSH drops, mounts stay in failed state forever.
New watchdog timer checks every 12min and restarts any failed mount.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

CLAUDE.md

@@ -73,12 +73,12 @@ miguel, pablo, patrick
 
 SFTP chroot: `/home/USER/html/`
 
-### ASIR1 (21) — usernames = apellido en minúsculas
+### ASIR1 (22) — usernames = apellido en minúsculas
 
 ```
 barja, barrios, cayo, contrera, duque, florea, gomes, izquierdo,
 jara, lillo, linares, macedo, martinez, munoz, olcina, ponce,
-posada, quiroz, reynoso, sierra, torrero
+posada, quiroz, reynoso, rodriguez, sierra, torrero
 ```
 
 SFTP chroot: `/home/USER/python/` · Contraseñas: leet-speak (`a→4, e→3, i→1, o→0`)

scripts/README.md

@@ -15,6 +15,8 @@ Copia de referencia de todos los scripts y unidades systemd desplegados en **zzz
 | **`nginx-user-config-watcher.service`** | `zzz:/etc/systemd/system/` | Unit systemd para el watcher DDAW2 |
 | **`home-fenix-napi-data.mount`** | `aldebaran:~/.config/systemd/user/` | Mount sshfs `~/napi-data/` → `zzz:/var/www/napi/data/` (DDAW2) |
 | **`home-fenix-napi-data2.mount`** | `aldebaran:~/.config/systemd/user/` | Mount sshfs `~/napi-data2/` → `zzz:/var/www/napi2/data/` (ASIR1) |
+| **`napi-mount-watchdog.service`** | `aldebaran:~/.config/systemd/user/` | Watchdog — reinicia mounts sshfs caídos |
+| **`napi-mount-watchdog.timer`** | `aldebaran:~/.config/systemd/user/` | Timer — ejecuta watchdog cada 12 min |
 
 ---
 
@@ -125,8 +127,20 @@ systemctl --user status home-fenix-napi\\x2ddata2.mount  # ASIR1
 # Montar/desmontar
 systemctl --user start home-fenix-napi\\x2ddata2.mount
 systemctl --user stop home-fenix-napi\\x2ddata2.mount
+
+# Watchdog (auto-recovery de mounts caídos)
+systemctl --user status napi-mount-watchdog.timer
+systemctl --user list-timers napi-mount-watchdog.timer
 ```
 
+### 🐕 napi-mount-watchdog (timer + service)
+
+**Propósito:** Auto-recuperación de mounts sshfs caídos. Si zzz se reinicia o la conexión SSH se corta, el watchdog detecta los mounts en estado `failed` y los reinicia automáticamente.
+
+- ⏱️ **Frecuencia:** cada 12 minutos (+ 30s tras boot)
+- 🔍 **Lógica:** comprueba `is-active` de ambos `.mount` → reinicia los que no estén `active`
+- 🛡️ **Complementa** el `reconnect` de sshfs (que solo maneja blips breves, no fallos de mount inicial)
+
 ---
 
 ## 🏗️ Estructura de directorios en zzz
@@ -166,3 +180,4 @@ systemctl --user stop home-fenix-napi\\x2ddata2.mount
 | `python-upload-watcher.sh` | **v7** | 2026-02-25 | Batching + re-entregas + Windows-safe + `__pycache__` filter + nombres completos |
 | `nginx-user-config-watcher.sh` | **v3** | 2026-02-19 | Multi-sitio + undeploy + CSP analysis |
 | `xmpp-notify.py` | **v1** | 2026-01-27 | Bot one-shot slixmpp |
+| `napi-mount-watchdog` | **v1** | 2026-03-10 | Timer+service watchdog para auto-recovery de mounts sshfs |

scripts/napi-mount-watchdog.service

@@ -0,0 +1,12 @@
+[Unit]
+Description=Restart failed napi sshfs mounts
+
+[Service]
+Type=oneshot
+ExecStart=/bin/bash -c '\
+  for unit in home-fenix-napi\\x2ddata.mount home-fenix-napi\\x2ddata2.mount; do \
+    state=$(systemctl --user is-active "$unit" 2>/dev/null); \
+    if [ "$state" != "active" ]; then \
+      systemctl --user restart "$unit"; \
+    fi; \
+  done'

scripts/napi-mount-watchdog.timer

@@ -0,0 +1,9 @@
+[Unit]
+Description=Check napi sshfs mounts every 2 minutes
+
+[Timer]
+OnBootSec=30
+OnUnitActiveSec=12min
+
+[Install]
+WantedBy=timers.target