devices(extreme): two 48-port SFP+ leaves carried copper cages #2284

Merged
jaylouisw merged 1 commit from catalogue/1120-0852 into main 2026-11-20 09:09:17 +00:00
Owner

No closing keyword: the two open device requests (#803 Huawei OptiXtrans E6616X, #130 Ubiquiti UDW-PRO) are still blocked on fields their vendors do not publish, so this pass's increment is the audit finding their threads left open.

What was wrong

node scripts/audit-port-types.js had a section of its own - Template ports at 10G+ presented as RJ45 copper on a box that describes itself as SFP - naming exactly two models, both Extreme, 48/48 template ports each:

- extreme/VSP 7254XSQ  48/48 template ports at 10G (1:1 -> rj45)
- extreme/X690-48x-2q-4c  48/48 template ports at 10G (1:1 -> rj45)

Neither spec declared a connector on its portTemplate, and 10G is not in the engine's optical rate list, so 96 fibre ports were drawn as copper sockets that would take a patch lead (the #60 shape, arriving through the template path).

Vendor evidence

Change

devices/extreme/x690-48x-2q-4c.json, devices/extreme/vsp-7254xsq.json: template gains connector: sfp+ and speedCaps: [1000, 10000]; the 40G ports gain qsfp+ / [10000, 40000]; the X690's four 100G ports gain qsfp28 / [10000, 25000, 40000, 50000, 100000]; both gain the sourceUrl and a notes field quoting the vendor line. 54 sockets per model before and after; no capability dropped. docs/CONNECTORS.md regenerated (973 -> 974 port shapes). Research note: research/extreme-48x-sfp-copper-cages.md.

Gates

  • node scripts/syntax-check.js - 486 files: all OK
  • node scripts/audit-port-types.js - offending devices 0, offending ports 0; the 10G+-as-RJ45 section is now empty; undeclared template ports 2,800 -> 2,704 in 126 specs
  • node scripts/run-tests.js - 406 suites, 13,481 assertions, 0 failed (incl. catalogue-hygiene, hardware-invariants, faceplate-catalogue, spec-vs-description, spec-survives-load)
  • node scripts/regen-registry-doc.js - 2,355 models, 29,735 sockets, 974 port shapes
No closing keyword: the two open device requests (#803 Huawei OptiXtrans E6616X, #130 Ubiquiti UDW-PRO) are still blocked on fields their vendors do not publish, so this pass's increment is the audit finding their threads left open. ## What was wrong `node scripts/audit-port-types.js` had a section of its own - *Template ports at 10G+ presented as RJ45 copper on a box that describes itself as SFP* - naming exactly two models, both Extreme, 48/48 template ports each: ``` - extreme/VSP 7254XSQ 48/48 template ports at 10G (1:1 -> rj45) - extreme/X690-48x-2q-4c 48/48 template ports at 10G (1:1 -> rj45) ``` Neither spec declared a `connector` on its portTemplate, and 10G is not in the engine's optical rate list, so 96 fibre ports were drawn as copper sockets that would take a patch lead (the #60 shape, arriving through the template path). ## Vendor evidence - **X690-48x-2q-4c (17350)** - *X690 Series Switch Quick Reference*, 121148-00, https://documentation.extremenetworks.com/HW_QRG/121148-00_x690_qrg.pdf (200, 801,340 B, text layer): "X690-48x-2q-4c Base unit, **48 1Gb/10Gb SFP+ ports**, 2 10Gb/40Gb QSFP+ ports, 4 10Gb/25Gb/40Gb/50Gb/100Gb QSFP28 ports"; front-panel legend: "1Gb/10GBASE-T ports (48t model) **or** 1Gb/10Gb SFP+ ports (48x model)". The copper twin is the separate part 17360. - **VSP 7254XSQ** - *VOSS 8.9 User Guide*, "VSP 7200 Series", https://documentation.extremenetworks.com/VOSS/SW/89/VOSSUserGuide/GUID-9364F72C-5FD5-4F86-ACCF-4D0C63C690E0.shtml (200, 1,317,315 B): "The **VSP 7254XSQ has 48 SFP/SFP+ fiber ports**. The VSP 7254XTQ has 48 RJ-45 copper ports." ## Change `devices/extreme/x690-48x-2q-4c.json`, `devices/extreme/vsp-7254xsq.json`: template gains `connector: sfp+` and `speedCaps: [1000, 10000]`; the 40G ports gain `qsfp+` / `[10000, 40000]`; the X690's four 100G ports gain `qsfp28` / `[10000, 25000, 40000, 50000, 100000]`; both gain the `sourceUrl` and a `notes` field quoting the vendor line. 54 sockets per model before and after; no capability dropped. `docs/CONNECTORS.md` regenerated (973 -> 974 port shapes). Research note: `research/extreme-48x-sfp-copper-cages.md`. ## Gates - `node scripts/syntax-check.js` - 486 files: all OK - `node scripts/audit-port-types.js` - offending devices 0, offending ports 0; the 10G+-as-RJ45 section is now **empty**; undeclared template ports 2,800 -> 2,704 in 126 specs - `node scripts/run-tests.js` - 406 suites, 13,481 assertions, 0 failed (incl. catalogue-hygiene, hardware-invariants, faceplate-catalogue, spec-vs-description, spec-survives-load) - `node scripts/regen-registry-doc.js` - 2,355 models, 29,735 sockets, 974 port shapes
devices(extreme): two 48-port SFP+ leaves carried copper cages
All checks were successful
CI / Classify the change (pull_request) Successful in 13s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 13s
CI / Server smoke and Docker build (pull_request) Successful in 23s
CI / Test (Node 20) (pull_request) Successful in 56s
CI / Test (Node 22) (pull_request) Successful in 56s
CI / Test (Node 18) (pull_request) Successful in 1m1s
CI / Arm auto-merge (pull_request) Successful in 4s
f14836c41b
The X690-48x-2q-4c and the VSP 7254XSQ declared no connector on their
portTemplate, and 10G is not in the engine's optical rate list, so
audit-port-types reported all 96 of their 10G ports as RJ45 copper on a
box that describes itself as SFP - the #60 shape through the template
path. Extreme states both are fibre: the X690 Quick Reference (121148-00)
says "48 1Gb/10Gb SFP+ ports" for the 48x model against "1Gb/10GBASE-T
ports (48t model)", and the VOSS 8.9 guide says "The VSP 7254XSQ has 48
SFP/SFP+ fiber ports. The VSP 7254XTQ has 48 RJ-45 copper ports."

Both specs now declare sfp+ on the template with 1G/10G speedCaps, qsfp+
on the 40G ports and qsfp28 on the X690's four 100G ports, and carry the
sourceUrl they were missing. Port counts and capabilities are unchanged;
docs/CONNECTORS.md regenerated (973 -> 974 port shapes). No issue is
closed: #803 and #130 stay blocked on unpublished vendor fields.
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-20 09:02:07 +00:00
jaylouisw deleted branch catalogue/1120-0852 2026-11-20 09:09:19 +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!2284
No description provided.