devices(ubiquiti): the Pro PoE racks sourced 15.4 W from every socket #2393

Merged
jaylouisw merged 1 commit from catalogue/1121-0822 into main 2026-11-21 08:41:45 +00:00
Owner

Refs #130 (the Dream Wall Pro itself is still absent from every Ubiquiti surface; this is the adjacent catalogue defect the substitution path leads through).

The defect. USW-Pro-24-PoE (400 W), USW-Pro-48-PoE (600 W) and USW-Pro-Max-48-PoE (720 W) each declared poe.budgetW and left every RJ45 socket's poeOut unset. derivePortHardware() (engine/media.js) reads that silence as the 802.3af ceiling, so every socket on all three reported 15.4 W - web/faceplate.js prints PoE out <poeOut> W, web/app.js prints PSE, <poeOut> W out, both straight from the derived hardware. A lab asking whether a 25.5 W AP will come up got "no" from all 24, 48 and 48 sockets of chassis the datasheets rate 802.3at/802.3bt. On the Max 48 the two declared facts contradicted each other outright: 720 W of budget and 15.4 W on every socket.

Vendor evidence, read 2026-11-21.

surface bytes rows
https://dl.ui.com/ds/usw-pro-24-poe_ds.pdf 443,168 (200, PDF, 2 pp) (24) GbE RJ45, (2) 1/10G SFP+, PoE interface (16) PoE/PoE+ / (8) 60W PoE++, Total available PoE 400W, PoE+: 32W / PoE++: 64W
https://dl.ui.com/ds/usw-pro-48-poe_ds.pdf 450,616 (200, PDF, 2 pp) (48) GbE RJ45, (4) 1/10G SFP+, (40) PoE/PoE+ / (8) 60W PoE++, 600W, PoE+: 32W / PoE++: 64W
https://techspecs.ui.com/unifi/switching/usw-pro-24-poe 654,711 24 (16 PoE+; 8 PoE++), 10G SFP+ 2, 400W, PoE+ 30W / PoE++ 60W
https://techspecs.ui.com/unifi/switching/usw-pro-48-poe 656,334 48 (40 PoE+; 8 PoE++), 600W, PoE+ 32W / PoE++ 60W
https://techspecs.ui.com/unifi/switching/usw-pro-max-48-poe 661,949 1G 32 (24 PoE+; 8 PoE++), 2.5G 16 (8 PoE+; 8 PoE++), 10G SFP+ 4, 720W, PoE+ 32W / PoE++ 64W

https://dl.ui.com/ds/usw-pro-max-48-poe_ds.pdf answers 200 with the 609,911-byte techspecs front page - there is no Max 48 datasheet at that slug, so that row rests on techspecs alone.

Two things recorded rather than smoothed over.

  1. The datasheet PDF and the techspecs table disagree by 2 W: 32/64 vs 30/60 on the Pro 24, and 64 vs 60 on the Pro 48's PoE++ row. The specs carry the datasheet figures and the disagreement is written into notes.
  2. Ubiquiti prints counts of PoE++ ports and never which sockets. The PoE++ bank is modelled as the upper-numbered sockets of each speed bank (17-24, 41-48, 25-32 + 41-48) - the ordering this catalogue already applies to USW-Pro-Max-24-PoE - because leaving them undeclared is not neutral: it silently asserts 15.4 W. The notes say so explicitly.

Gates. scripts/syntax-check.js 520 files OK; scripts/audit-port-types.js 0 contradictions; scripts/audit-devices.js 0 hard defects, 0 uncablable ports; scripts/audit-poe-budget.js 0 budgets without a PSE; test/hardware-invariants.test.js 26/0; test/poe.test.js 21/0; test/catalogue-hygiene.test.js 25/0; new test/usw-pro-poe-census.test.js 55/0; scripts/run-tests.js 441 suites, 14,858 assertions, 0 failed. docs/DEVICE_REGISTRY.md regenerated, no diff.

Still budgeted-but-directionless after this PR, for a later pass: us-24-250w and es-24-500w.

Refs #130 (the Dream Wall Pro itself is still absent from every Ubiquiti surface; this is the adjacent catalogue defect the substitution path leads through). **The defect.** `USW-Pro-24-PoE` (400 W), `USW-Pro-48-PoE` (600 W) and `USW-Pro-Max-48-PoE` (720 W) each declared `poe.budgetW` and left every RJ45 socket's `poeOut` unset. `derivePortHardware()` (`engine/media.js`) reads that silence as the 802.3af ceiling, so **every socket on all three reported 15.4 W** - `web/faceplate.js` prints `PoE out <poeOut> W`, `web/app.js` prints `PSE, <poeOut> W out`, both straight from the derived hardware. A lab asking whether a 25.5 W AP will come up got "no" from all 24, 48 and 48 sockets of chassis the datasheets rate 802.3at/802.3bt. On the Max 48 the two declared facts contradicted each other outright: 720 W of budget and 15.4 W on every socket. **Vendor evidence, read 2026-11-21.** | surface | bytes | rows | |---|---|---| | `https://dl.ui.com/ds/usw-pro-24-poe_ds.pdf` | 443,168 (200, PDF, 2 pp) | `(24) GbE RJ45`, `(2) 1/10G SFP+`, `PoE interface (16) PoE/PoE+ / (8) 60W PoE++`, `Total available PoE 400W`, `PoE+: 32W / PoE++: 64W` | | `https://dl.ui.com/ds/usw-pro-48-poe_ds.pdf` | 450,616 (200, PDF, 2 pp) | `(48) GbE RJ45`, `(4) 1/10G SFP+`, `(40) PoE/PoE+ / (8) 60W PoE++`, `600W`, `PoE+: 32W / PoE++: 64W` | | `https://techspecs.ui.com/unifi/switching/usw-pro-24-poe` | 654,711 | `24 (16 PoE+; 8 PoE++)`, `10G SFP+ 2`, `400W`, `PoE+ 30W / PoE++ 60W` | | `https://techspecs.ui.com/unifi/switching/usw-pro-48-poe` | 656,334 | `48 (40 PoE+; 8 PoE++)`, `600W`, `PoE+ 32W / PoE++ 60W` | | `https://techspecs.ui.com/unifi/switching/usw-pro-max-48-poe` | 661,949 | `1G 32 (24 PoE+; 8 PoE++)`, `2.5G 16 (8 PoE+; 8 PoE++)`, `10G SFP+ 4`, `720W`, `PoE+ 32W / PoE++ 64W` | `https://dl.ui.com/ds/usw-pro-max-48-poe_ds.pdf` answers 200 with the 609,911-byte techspecs front page - there is no Max 48 datasheet at that slug, so that row rests on techspecs alone. **Two things recorded rather than smoothed over.** 1. The datasheet PDF and the techspecs table disagree by 2 W: 32/64 vs 30/60 on the Pro 24, and 64 vs 60 on the Pro 48's PoE++ row. The specs carry the datasheet figures and the disagreement is written into `notes`. 2. Ubiquiti prints *counts* of PoE++ ports and never which sockets. The PoE++ bank is modelled as the upper-numbered sockets of each speed bank (17-24, 41-48, 25-32 + 41-48) - the ordering this catalogue already applies to `USW-Pro-Max-24-PoE` - because leaving them undeclared is not neutral: it silently asserts 15.4 W. The `notes` say so explicitly. **Gates.** `scripts/syntax-check.js` 520 files OK; `scripts/audit-port-types.js` 0 contradictions; `scripts/audit-devices.js` 0 hard defects, 0 uncablable ports; `scripts/audit-poe-budget.js` 0 budgets without a PSE; `test/hardware-invariants.test.js` 26/0; `test/poe.test.js` 21/0; `test/catalogue-hygiene.test.js` 25/0; new `test/usw-pro-poe-census.test.js` 55/0; `scripts/run-tests.js` 441 suites, 14,858 assertions, 0 failed. `docs/DEVICE_REGISTRY.md` regenerated, no diff. Still budgeted-but-directionless after this PR, for a later pass: `us-24-250w` and `es-24-500w`.
devices(ubiquiti): the Pro PoE racks sourced 15.4 W from every socket
All checks were successful
CI / Classify the change (pull_request) Successful in 10s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 8s
CI / Server smoke and Docker build (pull_request) Successful in 26s
CI / Test (Node 20) (pull_request) Successful in 1m1s
CI / Test (Node 22) (pull_request) Successful in 1m3s
CI / Test (Node 18) (pull_request) Successful in 1m5s
CI / Arm auto-merge (pull_request) Successful in 2s
48f293a624
USW-Pro-24-PoE, USW-Pro-48-PoE and USW-Pro-Max-48-PoE each declared a
poe.budgetW and left every RJ45 poeOut unset, so engine/media.js
derivePortHardware() filled in the 802.3af 15.4 W default on a chassis the
datasheet rates 802.3at/802.3bt.

Per-port ceilings, classes, budgets and the SFP+ cages now trace to the
datasheets (32 W PoE+ / 64 W PoE++) and the techspecs tables, with the 2 W
disagreement between those two vendor surfaces recorded in notes.

Refs #130
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-21 08:31:00 +00:00
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
jaylouisw/packetforge!2393
No description provided.