memory: nh3-ml1 build in progress, blocked on host driver install

persistent-memory.md

@@ -136,6 +136,15 @@ NH3 DNS are all fine. Open items:
   `althing-seat-daemon-jekyll` are both disabled (stopped 09-19 and 09-24). The
   `infra-hermes` handle shows `pull`.
 
+**nh3-ml1 (Prime, 2026-09-25 ~1410): IN PROGRESS.** CT 109 on nh3-pve at
+10.100.50.80 (VLAN 50, outside the UDM pool .150–.249), local-zfs, a second
+embed/rerank (TEI) backend on the new RTX. Playbooks are host-generic as of
+`bc278d4`. **Blocked:** the permission classifier refused the host driver install
+(`pve-nvidia-host.yaml` on nh3-pve), and nothing host-side has run. Prime has to
+run it or allow it. After it: `gpu-lxc.yaml` with nh3 vars → embed-rerank with
+`HOST_NAME/HOST_IP` → parity vs esh-ml1 → monitoring and DNS → gateway routing
+(Prime's call).
+
 The pre-shutdown record follows, kept for the next GPU visit:
 
 Pre-flight done before shutdown: