cli: show ip ospf interface brief is a view, not a port that is missing #2257

Merged
jaylouisw merged 1 commit from rev/1120-0412 into main 2026-11-20 04:39:14 +00:00
Owner

The brief keyword of show ip ospf interface was read as a port name and refused with % No interface brief on <box> - the sentence a misspelt port gets, about a command every IOS box has (the defect class #1443 fixed for show interfaces switchport).

Question (one, narrow). What does the brief view print - its cells, order and widths - and which of those cells can this engine fill from state it actually computes?

Verdict. Known for the shape, and the engine can fill four of seven cells:

  • Interface, Area, IP Address/Mask, Cost - all present in dev.ospfInterfaces.
  • PID - the process id from router ospf <process-id>, a command this console does not have (one unnamed process; the summary already says Routing Process with ID <router-id>). The Cisco Learning Network capture prints PID 3, so the cell is the operator's number, not a constant - printing 1 would claim a configuration nobody typed.
  • State - the port's DR/BDR role; computeOspf() elects a role only across a single cable with two speakers.
  • Nbrs F/C - the F half is dev.ospfNeighbors; the C half counts neighbours detected (2-WAY), a state the engine records as a refusal instead.

So the console now answers as a view: it names the view, prints the header at the vendor's measured widths, says the PID is the one cell no other view stands in for, and points at show ip ospf interface [<port>]. A port genuinely named brief still wins, because the port table is consulted first.

Sources. Cisco IOS OSPF Command Reference (ospf-s1.html, grammar and field list), the Academy Configuring Multiarea OSPFv2 lab PDF (the only capture whose spacing survives extraction - the widths are measured off it), Cisco Press "Implementing OSPFv2" (DR 0/0 on a lonely port), kazulog (a passive Gi2 keeps State DR at 0/0; F/C defined), Cisco Learning Network (PID 3), Ruckus FastIron command reference (the State vocabulary).

Changed.

  • engine/cli.js, _showOspf(): the interface branch tests the trailing word against the keyword's abbreviations (b, bri, brie, brief, any case) when the words are not a port, and answers with a four-line view refusal.
  • test/show-ip-ospf-interface-brief.test.js (new, 16 checks).
  • docs/CLI.md: the paragraph that documented % No interface brief on r1 as the console's answer.
  • docs/BGP-POLICY.md: two cli.js:7485 anchors moved to 7531 (the insert sits above _help()).
  • research/show-ip-ospf-interface-brief.md: the captures, the seven cells and their widths, what is derivable, and what would settle the rest.
  • CHANGELOG.md.

Gates. node scripts/syntax-check.js: 478 files, all OK. node scripts/check-doc-citations.js: no judged citation beyond drift. npm test: 398 suites, 13264 assertions passed, 0 failed.

Still unresolved. C in Nbrs F/C needs a 2-WAY neighbour (a heard-but-not-adjacent peer) to count, and the DROTHER spelling needs a third speaker on one segment; the PID waits on a modelling decision (router ospf <process-id>), not on evidence. What would settle the first two: a paste from three routers on one VLAN, and one from a Hello-interval mismatch.

The `brief` keyword of `show ip ospf interface` was read as a port name and refused with `% No interface brief on <box>` - the sentence a misspelt port gets, about a command every IOS box has (the defect class #1443 fixed for `show interfaces switchport`). **Question (one, narrow).** What does the `brief` view print - its cells, order and widths - and which of those cells can this engine fill from state it actually computes? **Verdict.** Known for the shape, and the engine can fill four of seven cells: - `Interface`, `Area`, `IP Address/Mask`, `Cost` - all present in `dev.ospfInterfaces`. - `PID` - the process id from `router ospf <process-id>`, a command this console does not have (one unnamed process; the summary already says `Routing Process with ID <router-id>`). The Cisco Learning Network capture prints `PID 3`, so the cell is the operator's number, not a constant - printing `1` would claim a configuration nobody typed. - `State` - the port's DR/BDR role; `computeOspf()` elects a role only across a single cable with two speakers. - `Nbrs F/C` - the `F` half is `dev.ospfNeighbors`; the `C` half counts neighbours *detected* (2-WAY), a state the engine records as a refusal instead. So the console now answers as a view: it names the view, prints the header at the vendor's measured widths, says the PID is the one cell no other view stands in for, and points at `show ip ospf interface [<port>]`. A port genuinely named `brief` still wins, because the port table is consulted first. **Sources.** Cisco IOS OSPF Command Reference (`ospf-s1.html`, grammar and field list), the Academy *Configuring Multiarea OSPFv2* lab PDF (the only capture whose spacing survives extraction - the widths are measured off it), Cisco Press "Implementing OSPFv2" (`DR 0/0` on a lonely port), kazulog (a passive `Gi2` keeps `State DR` at `0/0`; `F/C` defined), Cisco Learning Network (`PID 3`), Ruckus FastIron command reference (the State vocabulary). **Changed.** - `engine/cli.js`, `_showOspf()`: the `interface` branch tests the trailing word against the keyword's abbreviations (`b`, `bri`, `brie`, `brief`, any case) when the words are not a port, and answers with a four-line view refusal. - `test/show-ip-ospf-interface-brief.test.js` (new, 16 checks). - `docs/CLI.md`: the paragraph that documented `% No interface brief on r1` as the console's answer. - `docs/BGP-POLICY.md`: two `cli.js:7485` anchors moved to `7531` (the insert sits above `_help()`). - `research/show-ip-ospf-interface-brief.md`: the captures, the seven cells and their widths, what is derivable, and what would settle the rest. - `CHANGELOG.md`. **Gates.** `node scripts/syntax-check.js`: 478 files, all OK. `node scripts/check-doc-citations.js`: no judged citation beyond drift. `npm test`: 398 suites, 13264 assertions passed, 0 failed. **Still unresolved.** `C` in `Nbrs F/C` needs a 2-WAY neighbour (a heard-but-not-adjacent peer) to count, and the `DROTHER` spelling needs a third speaker on one segment; the PID waits on a modelling decision (`router ospf <process-id>`), not on evidence. What would settle the first two: a paste from three routers on one VLAN, and one from a Hello-interval mismatch.
cli: show ip ospf interface brief is a view, not a port that is missing
All checks were successful
CI / Classify the change (pull_request) Successful in 22s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 20s
CI / Server smoke and Docker build (pull_request) Successful in 30s
CI / Test (Node 22) (pull_request) Successful in 1m22s
CI / Test (Node 20) (pull_request) Successful in 1m25s
CI / Test (Node 18) (pull_request) Successful in 1m28s
CI / Arm auto-merge (pull_request) Successful in 3s
11494282be
The keyword was read as a port name and refused with the sentence a typo gets, about a command every IOS box has. It now names the view, prints the header at the vendor's measured widths, says which cell no other view stands in for (the PID from `router ospf <pid>`, which this console does not model) and points at `show ip ospf interface [<port>]`. The table is not printed: State is elected here only across a cable with two speakers and the C half of Nbrs F/C counts neighbours detected, which the engine does not record.

research/show-ip-ospf-interface-brief.md records the five captures, the seven cells with their widths, and which are derivable.
forgejo-actions scheduled this pull request to auto merge when all checks succeed 2026-11-20 04:30:17 +00:00
jaylouisw deleted branch rev/1120-0412 2026-11-20 04:39:24 +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!2257
No description provided.