devices(ubiquiti): the Gen1 PoE switches took their budget from the model name, and two of those names are wrong #2142

Merged
jaylouisw merged 1 commit from catalogue/1119-1122 into main 2026-11-19 11:40:59 +00:00
Owner

Refs #130

The Dream Wall Pro request is still blocked on a product Ubiquiti never listed (no udw-pro card in the cloud-gateway grid, 307 to the site root on techspecs, 404 on the store, no SWX-UDWPRO grant, no firmware platform code). Nothing was modelled from the non-Pro UDW. This PR is the catalogue defect found underneath the request: the Gen1 UniFi switches a routed lab is handed could not be trusted about how much PoE they can supply - and two of them were wrong.

The budget came from the model name. Eight devices/ubiquiti/us-* rows declared PSE ports with no poe.budgetW, so engine/device.js fell back to parsePoeBudget(model):

row name parse Ubiquiti "Total PoE Availability" now
us-8-60w.json 60 W 48 W (Port Layout 8 (4 PoE), PoE 15.4W/port) 48 W, cited
us-8-150w.json 150 W 130 W (8 (All PoE+)) 130 W, cited
us-24-250w.json 250 W 250 W (PoE 17W / PoE+ 34W per port) 250 W, cited
us-24-500w.json 500 W 500 W 500 W, cited
us-48-500w.json 500 W 500 W 500 W, cited
us-48-750w.json 750 W 750 W 750 W, cited

us-8-60w.json's own description already said 48 W while the allocator handed out 60 W, and a second Class-4 AP (25.5 W) fitted under 60 W but not under 48 W - which is the assertion the new test makes.

The same label over-reached the PSE count. media.js isPoePse() powers every copper port of a switch whose label claims PoE, and this label is "UniFi 8 PoE (Gen1)": ports 5-8 became 15.4 W PSEs on a model the vendor calls 8 (4 PoE) - 123.2 W of declared PSE capacity on a 48 W budget. Those four ports now say "poeOut": 0, the convention fortiswitch-124f-poe.json uses. No engine change.

Speeds. Every one of these pages prints the copper group as (1G/100M/10M); the rows offered only 1G/100M, so a 10M PoE phone could not negotiate on a port the vendor rates for it. 10M added to the copper ports of all eight rows.

Sources (all 200, server-rendered tables, read 2026-11-19): techspecs.ui.com/unifi/switching/{us-8-60w, us-8-150w, us-24, us-24-250w, us-24-500w, us-48, us-48-500w, us-48-750w}; 634,544 / 630,675 / 616,354 / 638,959 / 629,652 / 617,360 / 626,072 / 619,881 B. research/ubiquiti-unifi-gen1-poe-census.md carries the measurements and the traps: every dl.ui.com/ds/us-*_ds.pdf key is a 302 to the techspecs root, and us-16-150w has no page (307) and no store row (404), so its budget is left as it was rather than invented.

Gates: syntax-check 451 files OK; audit-port-types no contradiction; audit-poe-budget name-inferred 20 -> 14, "budget with no PSE port" 0; run-tests 372 suites, 12,659 assertions, 0 failed (new test/ubiquiti-gen1-poe-budget.test.js: 41).

Refs #130 The Dream Wall Pro request is still blocked on a product Ubiquiti never listed (no `udw-pro` card in the cloud-gateway grid, 307 to the site root on techspecs, 404 on the store, no `SWX-UDWPRO` grant, no firmware platform code). Nothing was modelled from the non-Pro UDW. This PR is the catalogue defect found underneath the request: the Gen1 UniFi switches a routed lab is handed could not be trusted about how much PoE they can supply - and two of them were wrong. **The budget came from the model name.** Eight `devices/ubiquiti/us-*` rows declared PSE ports with no `poe.budgetW`, so `engine/device.js` fell back to `parsePoeBudget(model)`: | row | name parse | Ubiquiti "Total PoE Availability" | now | |---|---|---|---| | `us-8-60w.json` | 60 W | **48 W** (`Port Layout 8 (4 PoE)`, `PoE 15.4W/port`) | 48 W, cited | | `us-8-150w.json` | 150 W | **130 W** (`8 (All PoE+)`) | 130 W, cited | | `us-24-250w.json` | 250 W | 250 W (`PoE 17W / PoE+ 34W` per port) | 250 W, cited | | `us-24-500w.json` | 500 W | 500 W | 500 W, cited | | `us-48-500w.json` | 500 W | 500 W | 500 W, cited | | `us-48-750w.json` | 750 W | 750 W | 750 W, cited | `us-8-60w.json`'s own description already said 48 W while the allocator handed out 60 W, and a second Class-4 AP (25.5 W) fitted under 60 W but not under 48 W - which is the assertion the new test makes. **The same label over-reached the PSE count.** `media.js` `isPoePse()` powers every copper port of a switch whose label claims PoE, and this label is "UniFi 8 PoE (Gen1)": ports 5-8 became 15.4 W PSEs on a model the vendor calls `8 (4 PoE)` - 123.2 W of declared PSE capacity on a 48 W budget. Those four ports now say `"poeOut": 0`, the convention `fortiswitch-124f-poe.json` uses. No engine change. **Speeds.** Every one of these pages prints the copper group as `(1G/100M/10M)`; the rows offered only `1G`/`100M`, so a 10M PoE phone could not negotiate on a port the vendor rates for it. `10M` added to the copper ports of all eight rows. Sources (all 200, server-rendered tables, read 2026-11-19): `techspecs.ui.com/unifi/switching/{us-8-60w, us-8-150w, us-24, us-24-250w, us-24-500w, us-48, us-48-500w, us-48-750w}`; 634,544 / 630,675 / 616,354 / 638,959 / 629,652 / 617,360 / 626,072 / 619,881 B. `research/ubiquiti-unifi-gen1-poe-census.md` carries the measurements and the traps: every `dl.ui.com/ds/us-*_ds.pdf` key is a 302 to the techspecs root, and `us-16-150w` has no page (307) and no store row (404), so its budget is left as it was rather than invented. Gates: `syntax-check` 451 files OK; `audit-port-types` no contradiction; `audit-poe-budget` name-inferred 20 -> 14, "budget with no PSE port" 0; `run-tests` 372 suites, 12,659 assertions, 0 failed (new `test/ubiquiti-gen1-poe-budget.test.js`: 41).
devices(ubiquiti): the Gen1 PoE switches took their budget from the model name, and two of those names are wrong (refs #130)
All checks were successful
CI / Classify the change (pull_request) Successful in 15s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 10s
CI / Server smoke test (pull_request) Successful in 11s
CI / Test (Node 20) (pull_request) Successful in 59s
CI / Test (Node 22) (pull_request) Successful in 58s
CI / Docker build (pull_request) Successful in 21s
CI / Test (Node 18) (pull_request) Successful in 1m4s
CI / Arm auto-merge (pull_request) Successful in 4s
2e85a087e1
Eight UniFi Gen1 switch rows declared PSE ports with no poe.budgetW, so engine/device.js
fell back to parsePoeBudget(model). US-8-60W was credited 60 W against Ubiquiti's
"Total PoE Availability = 48W", US-8-150W 150 W against 130 W - and us-8-60w.json's own
description already said 48 W. All six PoE rows now carry the vendor aggregate plus the
tech-spec URL; budgets are identical for the other four, traced instead of guessed.

The same label ("UniFi 8 PoE (Gen1)") made media.js isPoePse() power all eight copper
ports of the US-8-60W, whose vendor census is "8 (4 PoE)": the four data-only ports now
say poeOut: 0, as fortiswitch-124f-poe.json already does.

Copper speedCaps gain 10M on all eight rows: every page prints the group as (1G/100M/10M).
research/ubiquiti-unifi-gen1-poe-census.md records the measurements and the traps
(dl.ui.com datasheet keys 302 to the site root; us-16-150w has no page and is left alone).
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-19 11:36:15 +00:00
jaylouisw deleted branch catalogue/1119-1122 2026-11-19 11:41:02 +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!2142
No description provided.