huawei: NetEngine 8000 M4 verified against the vendor hardware description - MACsec is ports 0-19 only, AC ceiling is not the panel ceiling (#795) #1724

Merged
jaylouisw merged 1 commit from catalogue/1116-2025 into main 2026-11-16 20:29:09 +00:00
Owner

Fixes #795

What this settles

The NetEngine 8000 M4 has been in the catalogue since #1545, but #795 was never closed because that PR referenced the issue in its title instead of bodying Fixes #. This run re-verified every field of devices/huawei/netengine-8000-m4.json against the vendor's own hardware description and added the two facts the entry was missing.

Verification (source: https://support.huawei.com/enterprise/en/doc/EDOC1100465701/ed7ab88b/netengine-8000-m4, V800R024C10SPC500 Hardware Description)

  • Panel census confirmed - Table 1-4 "Ports on the NetEngine 8000 M4": 0 to 3 QSFP28 for 100GE/50GE/40GE (and the only ports with 4x10GE/4x25GE breakout), 4-19 SFP28 for 25GE/10GE/GE, 20-35 SFP28 for 25GE/10GE. 36 fixed cages, matching the entry exactly. The four PIC slots (AC: two) stay empty, as documented.
  • Capacity, both Huawei numbers - the hardware description's chassis tables say Switching capacity: 2.4 Tbit/s (Bidirectional); the M4 datasheet quotes the same box at 1.2 Tbit/s. Same fabric, counted once or per direction. The description now says 2.4 Tbit/s bidirectional and names the 1.2 Tbit/s reading instead of contradicting the notes.
  • DC and AC are not the same router above the panel - DC (CR8PM4BASDC2/DC6): 4 PICs, 116 ports, 12x100GE, 12x50GE, 64x25GE, 72x10GE, 96xGE, 283.4 W typical. AC (CR8PM4BASAC2/AC6): 2 PICs, still 116 ports, but 8x100GE, 8x50GE, 48x25GE, 52x10GE, 56xGE, 305.1 W typical. The fixed 36-cage panel is identical in both, so an AC chassis cannot reach the DC ceiling even though both are "116 ports".
  • MACsec stops at port 19 - every chassis table states MACsec supported: Yes. Ports 0 to 19 support MACsec. The GE/25GE SFP28 cages 4-19 carry it; the 25GE/10GE-only cages 20-35 and the four QSFP28 cages do not. 16 of the 36 cages have no MACsec, which the entry previously did not say.

Discrepancies stated

  • Datasheet 1.2 Tbit/s vs hardware description 2.4 Tbit/s bidirectional: both are Huawei, and the entry now shows the additive relationship rather than picking one silently.
  • The issue asks for mpls and bgp capabilities. Neither is an engine capability - the catalogue vocabulary is host, l2-forward, l3-forward, cli, dhcp-server, dhcp-client, nat, firewall, wireless, http-server, and 0 of the 2,329 specs declare either. Adding them would be an engine change, which is out of scope for this stream; the M4's L3 forwarding, NAT, firewall and CLI are declared.

Gates

  • node scripts/syntax-check.js - 360 files, all OK
  • node scripts/audit-port-types.js - "No port contradicts its declared type."
  • node scripts/regen-registry-doc.js - catalogue pages current: 2,329 models / 258 vendors / 28,413 sockets / 948 port shapes
  • node --test test/hardware-invariants.test.js test/catalogue-hygiene.test.js - 26 passed, 0 failed
Fixes #795 ## What this settles The NetEngine 8000 M4 has been in the catalogue since #1545, but #795 was never closed because that PR referenced the issue in its title instead of bodying `Fixes #`. This run re-verified every field of `devices/huawei/netengine-8000-m4.json` against the vendor's own hardware description and added the two facts the entry was missing. ## Verification (source: https://support.huawei.com/enterprise/en/doc/EDOC1100465701/ed7ab88b/netengine-8000-m4, V800R024C10SPC500 Hardware Description) - **Panel census confirmed** - Table 1-4 "Ports on the NetEngine 8000 M4": `0 to 3` QSFP28 for 100GE/50GE/40GE (and the only ports with 4x10GE/4x25GE breakout), `4-19` SFP28 for 25GE/10GE/GE, `20-35` SFP28 for 25GE/10GE. 36 fixed cages, matching the entry exactly. The four PIC slots (AC: two) stay empty, as documented. - **Capacity, both Huawei numbers** - the hardware description's chassis tables say `Switching capacity: 2.4 Tbit/s (Bidirectional)`; the M4 datasheet quotes the same box at `1.2 Tbit/s`. Same fabric, counted once or per direction. The description now says 2.4 Tbit/s bidirectional and names the 1.2 Tbit/s reading instead of contradicting the notes. - **DC and AC are not the same router above the panel** - DC (CR8PM4BASDC2/DC6): 4 PICs, 116 ports, 12x100GE, 12x50GE, 64x25GE, 72x10GE, 96xGE, 283.4 W typical. AC (CR8PM4BASAC2/AC6): 2 PICs, still 116 ports, but 8x100GE, 8x50GE, 48x25GE, 52x10GE, 56xGE, 305.1 W typical. The fixed 36-cage panel is identical in both, so an AC chassis cannot reach the DC ceiling even though both are "116 ports". - **MACsec stops at port 19** - every chassis table states `MACsec supported: Yes. Ports 0 to 19 support MACsec.` The GE/25GE SFP28 cages 4-19 carry it; the 25GE/10GE-only cages 20-35 and the four QSFP28 cages do not. 16 of the 36 cages have no MACsec, which the entry previously did not say. ## Discrepancies stated - Datasheet 1.2 Tbit/s vs hardware description 2.4 Tbit/s bidirectional: both are Huawei, and the entry now shows the additive relationship rather than picking one silently. - The issue asks for `mpls` and `bgp` capabilities. Neither is an engine capability - the catalogue vocabulary is `host, l2-forward, l3-forward, cli, dhcp-server, dhcp-client, nat, firewall, wireless, http-server`, and 0 of the 2,329 specs declare either. Adding them would be an engine change, which is out of scope for this stream; the M4's L3 forwarding, NAT, firewall and CLI are declared. ## Gates - `node scripts/syntax-check.js` - 360 files, all OK - `node scripts/audit-port-types.js` - "No port contradicts its declared type." - `node scripts/regen-registry-doc.js` - catalogue pages current: 2,329 models / 258 vendors / 28,413 sockets / 948 port shapes - `node --test test/hardware-invariants.test.js test/catalogue-hygiene.test.js` - 26 passed, 0 failed
huawei: NetEngine 8000 M4 - MACsec is ports 0-19 only, and the AC ceiling is not the panel ceiling (#795)
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 15s
CI / Server smoke test (pull_request) Has been skipped
CI / Docker build (pull_request) Has been skipped
CI / Test (Node 20) (pull_request) Successful in 53s
CI / Test (Node 18) (pull_request) Successful in 59s
CI / Test (Node 22) (pull_request) Successful in 59s
4b11522114
jaylouisw scheduled this pull request to auto merge when all checks succeed 2026-11-16 20:29:00 +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!1724
No description provided.