docs(hardware): record srv07 real specs and identity, correct mf04 mapping

srv07 verified on-site 2026-07-17: it is the Ansible/OS host mf04
(i3-4360T 2c/4t, 16 GB, 500 GB HDD, Debian 13), renamed to srv07 to match
the sticker. The naming-scheme migration row said mf04 -> srv05 — corrected
with a dated note. pp02: srv07's LAN path was restored 2026-07-17; the new
rear patch of port 8 is ad hoc and still needs recording.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

docs/guides/hardware-naming-scheme.md

@@ -99,7 +99,7 @@ Such hosts carry no `rack:` field and need no `cluster:`.
 | `mf01` | `srv02` | cluster omitted until a real assignment is given |
 | `mf02` | `srv03` | |
 | `mf03` | `srv04` | |
-| `mf04` | `srv05` | |
+| `mf04` | `srv07` | **corrected 2026-07-17** — originally recorded as `srv05`. Verified on-site: the physical box carrying the Ansible/OS host `mf04` (i3-4360T, 16 GB, 500 GB HDD) wears the `srv07` sticker; `srv05`'s page describes a different machine (i5-3570K). OS hostname renamed to `srv07` the same day. |
 | `pdu01`, `pdu02` | unchanged | already conform |
 | `sw01`, `pp01` | unchanged | already conform |
 | `makerfloss.eu` | unchanged | cloud FQDN exception |

docs/hardware/index.md

@@ -31,7 +31,7 @@ _Auto-generated from `docs/hardware/*.md` — do not edit by hand. Run `make doc
 | [srv04](srv04.md) | The pile | Intel Core i5-3570K @ 3.40GHz · 4c | 8 GB | 500 GB HDD | 1 GbE | staging |
 | [srv05](srv05.md) | The pile | Intel Core i5-3570K @ 3.40GHz · 4c | 8 GB | 500 GB HDD | 1 GbE | staging |
 | [srv06](srv06.md) | The pile | ? | ? | ? | ? | staging |
-| [srv07](srv07.md) | The pile | ? | ? | ? | ? | staging |
+| [srv07](srv07.md) | The pile | Intel Core i3-4360T @ 3.20GHz · 2c/4t | 16 GB | 500 GB HDD | 1 GbE | staging |
 
 ## Shelves
 

docs/hardware/pp02.md

@@ -23,6 +23,9 @@ links:
 - Port 1 → pp01:1 → wan01 (srv01 eth0, WAN).
 - Ports 2–7 → sw01:1–6 (LAN): srv01 eth1 (p2), srv02 (p3), srv03 (p4),
   srv04 (p5), srv05 (p6), srv06 (p7).
-- Port 8 still carries srv07 eth0, but its rear patch to sw01:7 was removed
+- Port 8 carries srv07 eth0. Its rear patch to sw01:7 was removed when sw01:7
-  when sw01:7 was repurposed as the uplink to sw06 — so srv07 (staging) has
+  was repurposed as the uplink to sw06, leaving srv07 with no LAN path for a
-  no active LAN path today.
+  while. **Connectivity was restored on 2026-07-17** (srv07 is back on the LAN
+  at 10.0.0.x and its wg1 tunnel is up) — but the new rear patching was done
+  ad hoc and is not yet recorded: confirm on-site which switch/port port 8 now
+  lands on and update this list.

docs/hardware/srv07.md

@@ -4,12 +4,13 @@ kind: server
 status: staging
 cluster: tappaas
 location: The pile
-cpu: "?"
+cpu: Intel Core i3-4360T @ 3.20GHz
-cpu_cores: "?"
+cpu_cores: 2
-cpu_threads: "?"
+cpu_threads: 4
-ram_gb: "?"
+ram_gb: 16
-storage: "?"
+storage_gb: 500
-nic_gbps: "?"
+storage_type: hdd
+nic_gbps: 1
 rack: rack01
 mounted_on: shf02
 shelf_face: front
@@ -24,4 +25,16 @@ links:
 
 ## Notes
 
-- Reconstructed from committed rack artifacts; specs are placeholders.
+- Specs read live 2026-07-17 (lscpu/free/lsblk): i3-4360T (2c/4t), 16 GB RAM,
+  500 GB HDD (Seagate ST3500418AS). Runs Debian 13 (trixie).
+- This is the machine Ansible (AnsibleBaobabV4) managed as **mf04**; the OS
+  hostname was renamed `mf04` → `srv07` on 2026-07-17 to match the sticker.
+  The naming-scheme guide's migration row originally said `mf04` → `srv05`;
+  that was a mis-mapping (srv05's page describes an i5-3570K box).
+- SSH on **:7576** (project standard), user `sjat`. WireGuard split-tunnel to
+  the makerfloss VPS (`wg1` peer `10.13.0.3`, unit `wg-quick@mf-srv07`) dials
+  out automatically on boot.
+- LAN path restored 2026-07-17 (DHCP lease `10.0.0.183`). The rear patch
+  pp02:8 → sw01:7 had been removed when sw01:7 became the sw06 uplink, so the
+  box had no LAN path; the current rear patching still needs to be recorded —
+  confirm on-site which switch/port it lands on and update this page + pp02.