research(trunk): the Status column is not-trunking when the link is up, other when it is not #2190

Merged
jaylouisw merged 1 commit from rev/1119-1812 into main 2026-11-19 18:38:22 +00:00
Owner

Question

show interfaces <port> trunk section 1 has a Status column. When the named port is not trunking, does it carry not-trunking (Catalyst 9600 17.7 reference) or other (3850/9200/9300 references)? research/show-interfaces-trunk-shape.md left this contradictory and blocked step 2 of docs/plans/interfaces-trunk-view.md on it.

Verdict

The two words partition port states, they do not compete:

  • trunking - the port is an operational trunk.
  • not-trunking - the link is up and the port is not trunking (static access, off, or a dynamic mode that did not negotiate).
  • other - the link is not up (no cable, notconnect), so there is no operational trunking state to report.

Five real-box captures fit this: Cat 3550 trunk/down -> other (CLN); Cat 2950 static access/up -> off + not-trunking, desirable/up facing access -> not-trunking (Cisco Community native-VLAN-mismatch thread); cabled trunks -> trunking; Cat 6500 ISL guide desirable/up -> not-trunking. The 9600/9300 pair is two different port states, not two vendors' opinions - the 9600 sample forwards VLAN 1 in the last block, the 9300 sample is none everywhere.

Two mechanisms still fit every capture (link-up vs administrative mode). No capture separates them, and they agree on every state this engine can reach (no DTP). The note records both, names the capture that would decide it, and lists the still-open cases (administratively down, err-disabled, up + configured trunk + not trunking).

Scope

Research only, no engine code. The view does not exist and step 2 is a new view that needs approval. This settles rule 7 of the shape note and both documents now point at the new note.

Files

  • research/show-interfaces-trunk-status-word.md (new)
  • research/show-interfaces-trunk-shape.md - verdict paragraph, open-questions row and rule 7 now carry the settled answer
  • docs/plans/interfaces-trunk-view.md - the Status risk points at the note instead of asking step 1 to decide

Gates

  • node scripts/syntax-check.js: 461 files OK
  • node scripts/check-control-bytes.js: clean
  • node scripts/check-doc-citations.js: no judged citation beyond 40 lines
  • npm test: 382 suites, 12,890 assertions, 0 failed
## Question `show interfaces <port> trunk` section 1 has a `Status` column. When the named port is not trunking, does it carry `not-trunking` (Catalyst 9600 17.7 reference) or `other` (3850/9200/9300 references)? `research/show-interfaces-trunk-shape.md` left this **contradictory** and blocked step 2 of `docs/plans/interfaces-trunk-view.md` on it. ## Verdict The two words partition port states, they do not compete: - `trunking` - the port is an operational trunk. - `not-trunking` - the link is **up** and the port is not trunking (static access, `off`, or a dynamic mode that did not negotiate). - `other` - the link is **not up** (no cable, notconnect), so there is no operational trunking state to report. Five real-box captures fit this: Cat 3550 `trunk`/down -> `other` (CLN); Cat 2950 static access/up -> `off` + `not-trunking`, `desirable`/up facing access -> `not-trunking` (Cisco Community native-VLAN-mismatch thread); cabled trunks -> `trunking`; Cat 6500 ISL guide `desirable`/up -> `not-trunking`. The 9600/9300 pair is two different port states, not two vendors' opinions - the 9600 sample forwards VLAN 1 in the last block, the 9300 sample is `none` everywhere. Two mechanisms still fit every capture (link-up vs administrative mode). No capture separates them, and they agree on every state this engine can reach (no DTP). The note records both, names the capture that would decide it, and lists the still-open cases (administratively down, err-disabled, up + configured trunk + not trunking). ## Scope Research only, no engine code. The view does not exist and step 2 is a new view that needs approval. This settles rule 7 of the shape note and both documents now point at the new note. ## Files - `research/show-interfaces-trunk-status-word.md` (new) - `research/show-interfaces-trunk-shape.md` - verdict paragraph, open-questions row and rule 7 now carry the settled answer - `docs/plans/interfaces-trunk-view.md` - the Status risk points at the note instead of asking step 1 to decide ## Gates - `node scripts/syntax-check.js`: 461 files OK - `node scripts/check-control-bytes.js`: clean - `node scripts/check-doc-citations.js`: no judged citation beyond 40 lines - `npm test`: 382 suites, 12,890 assertions, 0 failed
research(trunk): the Status column is not-trunking when the link is up, other when it is not
All checks were successful
CI / Classify the change (pull_request) Successful in 31s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 12s
CI / Server smoke test (pull_request) Successful in 11s
CI / Test (Node 22) (pull_request) Successful in 1m10s
CI / Test (Node 20) (pull_request) Successful in 1m12s
CI / Test (Node 18) (pull_request) Successful in 1m16s
CI / Docker build (pull_request) Successful in 20s
CI / Arm auto-merge (pull_request) Successful in 3s
760ba6b092
`show interfaces <port> trunk` prints one of three words in its Status
column, and research/show-interfaces-trunk-shape.md left two of them
contradictory: the Catalyst 9600 17.7 reference prints `not-trunking`
where the 3850/9200/9300 references print `other`.

Five captures from real boxes show the two words are not rival spellings
of one state. A configured trunk with no cable (Cat 3550, mode `trunk`,
down/down) reads `other`; a static access port that is up (Cat 2950,
`off` in the Mode column) reads `not-trunking`; a `desirable` port whose
peer is access, link up, reads `not-trunking`; a cabled trunk reads
`trunking`. The 9600/9300 pair is two different port states, not two
vendors' opinions: the 9600 sample forwards VLAN 1, the 9300 sample has
`none` in all three VLAN blocks.

Two mechanisms fit every capture (link-up vs administrative-mode), and
no capture separates them, so the note records both and the one state
that would decide it. They agree on every state this engine can reach,
because there is no DTP here.

No engine code changes: the view does not exist and step 2 of
docs/plans/interfaces-trunk-view.md is a new view that needs approval.
The note settles the rule step 2 was blocked on (rule 7 of the shape
note), and both documents now point at it.

Sources (in the note): CLN "show interface trunk status other"; Cisco
Community "Native VLAN mismatch discovered on switch"; Cisco ISL support
document 10682-2; the 9300 17.15 / 9600 17.7 command references; the
IOS XE interface command reference.
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-19 18:21:58 +00:00
jaylouisw deleted branch rev/1119-1812 2026-11-19 18:38:24 +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!2190
No description provided.