Netonix: nine WISP switches with per-port passive PoE, real budgets and 115200 consoles #1560

Merged
jaylouisw merged 2 commits from catalogue/1115-2256 into main 2026-11-16 01:35:12 +00:00
Owner

What this adds

The catalogue had exactly one Netonix model, WS-12-250-AC, and it declared no PoE: no poe
block, no poeOut on any port, no poe-pse capability. The simulator therefore believed a
12-port passive-PoE WISP switch supplied no power at all. This PR fills in the family.

Added (8): WS-12-250-DC, WS-12-400-AC, WS-12-DC, WS-26-400-AC, WS-26-400-IDC,
WS-26-500-DC, WS-8-150-AC, WS-6-MINI
Corrected (1): WS-12-250-AC now carries its 250 W ceiling, a PSE flag on all 12 copper
ports, the sfp cages as cages, the console port, the fabric figure and the notes the other
members carry.

Sources

source used for
WISP Switch spec sheet / quick start dimensions, weight, power input, max consumption, line rate, PoE pinout and current limits, CPU/RAM/flash, temperature, PoE port ranges, console settings, default management IP
https://netonix.com/<model>.html (one per model, full table in each spec's sourceUrl and poe.sourceUrl) port counts as the vendor sells them, supply rating, the tiered PoE budget on the DC models, the WS-6-MINI's power-in rules

Every field in these specs carries the URL it came from; the notes string names the PDF where
a fact came from it and the store page where it came from that.

PoE modelling, and why there is no wattage class

These are not 802.3 PSEs. They source software-selectable passive PoE: 24 V and 48 V at 1 A on
pins 4,5(+)/7,8(-), 24VH and 48VH at 2 A on all four pairs, with a current sensor per port. A
port's wattage is whatever voltage and current limit a human picks, so poeOut is true (this
port sources power) rather than a class number, and the numbers that do appear are the vendor's
own ceilings:

  • WS-12-250-AC, WS-12-250-DC: 250 W max consumption. The DC model's store page is explicit
    that the budget tracks input voltage: 250 W above 12 V in, 200 W above 11 V, 150 W below 10 V.
  • WS-12-400-AC: 400 W supply rating (the sheet prints N/A for max consumption).
  • WS-12-DC: budgetUnknown. Its 48 V output is a pass-through of the 37-54 VDC input and the
    unit only draws 4-10 W itself, so any number would be an invention about someone else's supply.
  • WS-26-400-AC / -IDC: 400 W, WS-26-500-DC: 500 W, WS-8-150-AC: 150 W,
    WS-6-MINI: 100 W (of which the switch takes 4 W).

WS-6-MINI is the interesting one: it is a powered device as well as a PSE. Port 1 takes
passive 48VH on all four pairs (48-56 V, up to 2 A) from another WISP switch, or a 48-54 V
barrel supply, and no other port sources power off a 2-pair 48 V adapter. It declares
poe-pd with poeIn: 100 on port 1 and poeOut on ports 2-6.

Discrepancies, stated not smoothed

  • Netonix's "Total Non-Blocking Line Rate" is not a duplex figure. 14 Gbps is 12x1G + 2x1G
    summed one-directionally, 26 Gbps is 24x1G + 2x1G. A FortiSwitch in this same catalogue calls
    24x1G + 4x10G "128 Gbps" by doubling. Like for like, these Netonix numbers would be 28 and 52.
    The fabric.note on each spec says so, because a bare comparison against other vendors'
    "switching capacity" would be wrong by 2x in the opposite direction.
  • WS-8-150-AC's spec sheet claims 14 Gbps of line rate for eight gigabit ports. That is the
    12-port row copied down: eight gigabit ports cannot exceed 8 Gbps. The port count itself is not
    in doubt (store page, PoE table and faceplate art all say 8, no SFP), so the spec is included
    with no fabric block rather than with a figure the sheet cannot support.
  • Two models were excluded because the vendor contradicts itself, and a guess would be the
    one thing the evidence rule forbids:
    • WS-26-DC: store page says "Managed passive POE Switch (12) Gbps POE Ports (2) SFP";
      the spec sheet's PoE table says ports 01-24 and its faceplate art shows 1/2 through 23/24
      plus 25/26. 12+2 and 24+2 cannot both be true.
    • WS-8-150-DC: store page says "Managed 6 Port POE Switch + 2 SFP Uplink Ports" and its PoE
      table says ports 01-06; the spec sheet says "(3) 10/100/1000 Ethernet Ports (2) SFP" and its
      PoE table says ports 01-08.
      Both need a vendor photo or a unit in hand before they enter the catalogue.

Validation

node scripts/syntax-check.js            -> Syntax-checked 325 files: all OK
node scripts/audit-port-types.js        -> 2082 models across 252 vendors, 21682 ports,
                                           offending devices: 0, offending ports: 0   (exit 0)
node scripts/check-control-bytes.js     -> No control bytes in source.
node scripts/import-devices.js --json   -> added 8, warnings 0 (validated, then --apply)
node test/hardware-invariants.test.js   -> pass
node test/poe.test.js                   -> pass
node test/catalogue-hygiene.test.js     -> pass
node test/import-devices-capabilities.test.js -> pass  (the capability tripwire)
node test/topology-roundtrip.test.js    -> pass
node test/doc-catalog-counts.test.js    -> 24/24 after `regen:registry-doc`
node test/faceplate-catalogue.test.js   -> pass
node test/apply-research.test.js        -> pass

End-to-end PoE check, because a family that suddenly sources power should be seen doing it:

TOWER (WS-26-400-AC) --Port 7--> Port 1 (WS-6-MINI)

show power inline                    Total: 100 W of 400 W in use, 300 W left, nothing denied
budget capped to 60 W ->             Port 7 ... off ... (denied)
                                     "PSE budget exhausted: 100 W requested, 60 W left"

That is the WISP pattern these boxes exist for: one tower switch feeding a small PoE-powered
switch up the mast.

Unverifiable / left alone

  • No firmware revision, bootloader or UI version is claimed: Netonix publishes none on these
    pages, and the spec sheet is undated beyond its 2018 copyright.
  • WS-26-DC and WS-8-150-DC, as above.
  • The store pages were read through the Wayback Machine for four models (the live site 403s a
    burst of requests); the archived snapshots are 2023-2026 and are the same copy as the live
    page where both were readable.
## What this adds The catalogue had exactly one Netonix model, `WS-12-250-AC`, and it declared no PoE: no `poe` block, no `poeOut` on any port, no `poe-pse` capability. The simulator therefore believed a 12-port passive-PoE WISP switch supplied no power at all. This PR fills in the family. **Added (8):** `WS-12-250-DC`, `WS-12-400-AC`, `WS-12-DC`, `WS-26-400-AC`, `WS-26-400-IDC`, `WS-26-500-DC`, `WS-8-150-AC`, `WS-6-MINI` **Corrected (1):** `WS-12-250-AC` now carries its 250 W ceiling, a PSE flag on all 12 copper ports, the `sfp` cages as cages, the console port, the fabric figure and the notes the other members carry. ## Sources | source | used for | |---|---| | [WISP Switch spec sheet / quick start](https://netonix.com/media/wysiwyg/ws-specsheet.pdf) | dimensions, weight, power input, max consumption, line rate, PoE pinout and current limits, CPU/RAM/flash, temperature, PoE port ranges, console settings, default management IP | | `https://netonix.com/<model>.html` (one per model, full table in each spec's `sourceUrl` and `poe.sourceUrl`) | port counts as the vendor sells them, supply rating, the tiered PoE budget on the DC models, the WS-6-MINI's power-in rules | Every field in these specs carries the URL it came from; the `notes` string names the PDF where a fact came from it and the store page where it came from that. ## PoE modelling, and why there is no wattage class These are not 802.3 PSEs. They source software-selectable *passive* PoE: 24 V and 48 V at 1 A on pins 4,5(+)/7,8(-), 24VH and 48VH at 2 A on all four pairs, with a current sensor per port. A port's wattage is whatever voltage and current limit a human picks, so `poeOut` is `true` (this port sources power) rather than a class number, and the numbers that do appear are the vendor's own ceilings: - `WS-12-250-AC`, `WS-12-250-DC`: 250 W max consumption. The DC model's store page is explicit that the budget tracks input voltage: 250 W above 12 V in, 200 W above 11 V, 150 W below 10 V. - `WS-12-400-AC`: 400 W supply rating (the sheet prints `N/A` for max consumption). - `WS-12-DC`: `budgetUnknown`. Its 48 V output is a pass-through of the 37-54 VDC input and the unit only draws 4-10 W itself, so any number would be an invention about someone else's supply. - `WS-26-400-AC` / `-IDC`: 400 W, `WS-26-500-DC`: 500 W, `WS-8-150-AC`: 150 W, `WS-6-MINI`: 100 W (of which the switch takes 4 W). `WS-6-MINI` is the interesting one: it is a powered device as well as a PSE. Port 1 takes passive **48VH on all four pairs** (48-56 V, up to 2 A) from another WISP switch, or a 48-54 V barrel supply, and no other port sources power off a 2-pair 48 V adapter. It declares `poe-pd` with `poeIn: 100` on port 1 and `poeOut` on ports 2-6. ## Discrepancies, stated not smoothed - **Netonix's "Total Non-Blocking Line Rate" is not a duplex figure.** 14 Gbps is 12x1G + 2x1G summed one-directionally, 26 Gbps is 24x1G + 2x1G. A FortiSwitch in this same catalogue calls 24x1G + 4x10G "128 Gbps" by doubling. Like for like, these Netonix numbers would be 28 and 52. The `fabric.note` on each spec says so, because a bare comparison against other vendors' "switching capacity" would be wrong by 2x in the opposite direction. - **`WS-8-150-AC`'s spec sheet claims 14 Gbps of line rate for eight gigabit ports.** That is the 12-port row copied down: eight gigabit ports cannot exceed 8 Gbps. The port count itself is not in doubt (store page, PoE table and faceplate art all say 8, no SFP), so the spec is included with **no** `fabric` block rather than with a figure the sheet cannot support. - **Two models were excluded because the vendor contradicts itself**, and a guess would be the one thing the evidence rule forbids: - `WS-26-DC`: store page says "Managed passive POE Switch (12) Gbps POE Ports (2) SFP"; the spec sheet's PoE table says ports 01-24 and its faceplate art shows 1/2 through 23/24 plus 25/26. 12+2 and 24+2 cannot both be true. - `WS-8-150-DC`: store page says "Managed 6 Port POE Switch + 2 SFP Uplink Ports" and its PoE table says ports 01-06; the spec sheet says "(3) 10/100/1000 Ethernet Ports (2) SFP" and its PoE table says ports 01-08. Both need a vendor photo or a unit in hand before they enter the catalogue. ## Validation ``` node scripts/syntax-check.js -> Syntax-checked 325 files: all OK node scripts/audit-port-types.js -> 2082 models across 252 vendors, 21682 ports, offending devices: 0, offending ports: 0 (exit 0) node scripts/check-control-bytes.js -> No control bytes in source. node scripts/import-devices.js --json -> added 8, warnings 0 (validated, then --apply) node test/hardware-invariants.test.js -> pass node test/poe.test.js -> pass node test/catalogue-hygiene.test.js -> pass node test/import-devices-capabilities.test.js -> pass (the capability tripwire) node test/topology-roundtrip.test.js -> pass node test/doc-catalog-counts.test.js -> 24/24 after `regen:registry-doc` node test/faceplate-catalogue.test.js -> pass node test/apply-research.test.js -> pass ``` End-to-end PoE check, because a family that suddenly sources power should be seen doing it: ``` TOWER (WS-26-400-AC) --Port 7--> Port 1 (WS-6-MINI) show power inline Total: 100 W of 400 W in use, 300 W left, nothing denied budget capped to 60 W -> Port 7 ... off ... (denied) "PSE budget exhausted: 100 W requested, 60 W left" ``` That is the WISP pattern these boxes exist for: one tower switch feeding a small PoE-powered switch up the mast. ## Unverifiable / left alone - No firmware revision, bootloader or UI version is claimed: Netonix publishes none on these pages, and the spec sheet is undated beyond its 2018 copyright. - `WS-26-DC` and `WS-8-150-DC`, as above. - The store pages were read through the Wayback Machine for four models (the live site 403s a burst of requests); the archived snapshots are 2023-2026 and are the same copy as the live page where both were readable.
netonix: nine WISP switches with per-port passive PoE, real budgets and 115200 consoles
All checks were successful
CI / Classify the change (pull_request) Successful in 19s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 22s
CI / Test (Node 20) (pull_request) Successful in 3m12s
CI / Test (Node 18) (pull_request) Successful in 3m24s
CI / Server smoke test (pull_request) Successful in 33s
CI / Docker build (pull_request) Successful in 23s
CI / Test (Node 22) (pull_request) Successful in 3m8s
d1cdf2ac8f
The catalogue had one Netonix model (WS-12-250-AC) and it carried no PoE at
all, so the simulator treated a 12-port passive-PoE switch as a box that
supplies no power. This fills in the family from the vendor's own WISP Switch
spec sheet plus each model's store page.

Added: WS-12-250-DC, WS-12-400-AC, WS-12-DC, WS-26-400-AC, WS-26-400-IDC,
WS-26-500-DC, WS-8-150-AC, WS-6-MINI. Corrected: WS-12-250-AC.

Every port, budget, dimension and console setting carries the URL it came
from. PoE is modelled as poe-pse with a per-port poeOut flag, not an 802.3
class: these boxes source software-selectable 24V/48V/24VH/48VH passive PoE
at 1 A or 2 A per port, so a wattage class would be an invention.

Excluded, and said so rather than guessed: WS-26-DC and WS-8-150-DC, whose
port counts contradict themselves across the vendor's own pages.
jaylouisw force-pushed catalogue/1115-2256 from d1cdf2ac8f
All checks were successful
CI / Classify the change (pull_request) Successful in 19s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 22s
CI / Test (Node 20) (pull_request) Successful in 3m12s
CI / Test (Node 18) (pull_request) Successful in 3m24s
CI / Server smoke test (pull_request) Successful in 33s
CI / Docker build (pull_request) Successful in 23s
CI / Test (Node 22) (pull_request) Successful in 3m8s
to d16549f609
Some checks failed
CI / Classify the change (pull_request) Successful in 22s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 17s
CI / Test (Node 18) (pull_request) Failing after 3m28s
CI / Test (Node 20) (pull_request) Failing after 3m11s
CI / Server smoke test (pull_request) Has been skipped
CI / Docker build (pull_request) Has been skipped
CI / Test (Node 22) (pull_request) Failing after 3m4s
2026-11-15 23:26:11 +00:00
Compare
Author
Owner

This branch conflicts with main, so it was not rebased:

devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json 

A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.

This branch conflicts with `main`, so it was not rebased: ``` devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json ``` A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.
Author
Owner

This branch conflicts with main, so it was not rebased:

devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json 

A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.

This branch conflicts with `main`, so it was not rebased: ``` devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json ``` A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.
jaylouisw force-pushed catalogue/1115-2256 from d16549f609
Some checks failed
CI / Classify the change (pull_request) Successful in 22s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 17s
CI / Test (Node 18) (pull_request) Failing after 3m28s
CI / Test (Node 20) (pull_request) Failing after 3m11s
CI / Server smoke test (pull_request) Has been skipped
CI / Docker build (pull_request) Has been skipped
CI / Test (Node 22) (pull_request) Failing after 3m4s
to 20ec902a7a
Some checks failed
CI / Classify the change (pull_request) Successful in 13s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 20s
CI / Test (Node 18) (pull_request) Failing after 49s
CI / Server smoke test (pull_request) Has been skipped
CI / Test (Node 20) (pull_request) Failing after 59s
CI / Docker build (pull_request) Has been skipped
CI / Test (Node 22) (pull_request) Failing after 4m0s
2026-11-16 01:11:22 +00:00
Compare
jaylouisw force-pushed catalogue/1115-2256 from 20ec902a7a
Some checks failed
CI / Classify the change (pull_request) Successful in 13s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 20s
CI / Test (Node 18) (pull_request) Failing after 49s
CI / Server smoke test (pull_request) Has been skipped
CI / Test (Node 20) (pull_request) Failing after 59s
CI / Docker build (pull_request) Has been skipped
CI / Test (Node 22) (pull_request) Failing after 4m0s
to e2b3fd1201
Some checks failed
CI / Classify the change (pull_request) Successful in 24s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 24s
CI / Test (Node 18) (pull_request) Failing after 32s
CI / Server smoke test (pull_request) Has been skipped
CI / Test (Node 22) (pull_request) Failing after 3m34s
CI / Test (Node 20) (pull_request) Failing after 3m41s
CI / Docker build (pull_request) Failing after 12m30s
2026-11-16 01:34:09 +00:00
Compare
Author
Owner

This branch conflicts with main, so it was not rebased:

devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json 

A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.

This branch conflicts with `main`, so it was not rebased: ``` devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json ``` A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.
Author
Owner

This branch conflicts with main, so it was not rebased:

devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json 

A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.

This branch conflicts with `main`, so it was not rebased: ``` devices/netonix/ws-12-250-ac.json devices/netonix/ws-12-250-dc.json devices/netonix/ws-12-400-ac.json devices/netonix/ws-12-dc.json devices/netonix/ws-26-400-ac.json devices/netonix/ws-26-400-idc.json devices/netonix/ws-26-500-dc.json devices/netonix/ws-6-mini.json devices/netonix/ws-8-150-ac.json docs/DEVICE_REGISTRY.md research/incoming/devices-netonix-wisp.json ``` A conflict is a decision about behaviour rather than a mechanical fix, so this is left for a human. Rebasing by hand and force-pushing will let the automation pick it up again.
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!1560
No description provided.