A no channel-group undoes the group it names, and a port belongs to one bundle (#1510) #1543

Merged
jaylouisw merged 1 commit from fix/1115-2035 into main 2026-11-15 21:05:43 +00:00
Owner

Fixes #1510

no channel-group <id> threw the id away: the dispatcher called _channelGroup(['remove'], iface), so
no channel-group 9999 answered Fa0/1 removed from channel group 2 (the group had no other members) and the
port left the bundle that was carrying traffic.

What changed in engine/cli.js:

  • the id is read. no channel-group <id> validates it (1-4096, both ends - the shared
    _channelGroupId() is now used by the set form too, which tested only the lower bound while its refusal
    stated the whole range) and removes the port from that group. An id that is not the port's group is a
    refusal naming both groups: % Fa0/1 is not in channel group 5: it is in group 2. A bare
    no channel-group keeps its old meaning - whatever the port is in.
  • a port belongs to one bundle. A second channel-group moves it, and says
    Interface Fa0/1 moved from channel group 2 to 10 (mode on); the group it left is deleted if it is now
    empty.
  • show etherchannel summary orders bundles by number (2, 10, 100), not as strings (10, 100, 2).

docs/CLI.md gains a short section next to the etherchannel table.

Tests: test/no-command-touches-only-what-it-names.test.js - 4 new checks, 11/11 in the file; three of them
fail against the previous handler (the wrong-group no, the move, the order). Full suite 253 suites / 9243
assertions / 0 failed
; syntax-check 323 files; lag-mode 50/50, lag-member-checks 8/8,
lag-member-down 30/30, l2 28/28.

Not done here: #1505's own test (stated-range-is-the-enforced-range.test.js) - the ceiling it asks for falls
out of the shared check, but the four-point probe table is left to that issue.

Fixes #1510 `no channel-group <id>` threw the id away: the dispatcher called `_channelGroup(['remove'], iface)`, so `no channel-group 9999` answered `Fa0/1 removed from channel group 2 (the group had no other members)` and the port left the bundle that was carrying traffic. What changed in `engine/cli.js`: - **the id is read.** `no channel-group <id>` validates it (`1-4096`, both ends - the shared `_channelGroupId()` is now used by the set form too, which tested only the lower bound while its refusal stated the whole range) and removes the port from **that** group. An id that is not the port's group is a refusal naming both groups: `% Fa0/1 is not in channel group 5: it is in group 2`. A bare `no channel-group` keeps its old meaning - whatever the port is in. - **a port belongs to one bundle.** A second `channel-group` moves it, and says `Interface Fa0/1 moved from channel group 2 to 10 (mode on)`; the group it left is deleted if it is now empty. - **`show etherchannel summary` orders bundles by number** (`2, 10, 100`), not as strings (`10, 100, 2`). `docs/CLI.md` gains a short section next to the etherchannel table. Tests: `test/no-command-touches-only-what-it-names.test.js` - 4 new checks, **11/11** in the file; three of them fail against the previous handler (the wrong-group `no`, the move, the order). Full suite **253 suites / 9243 assertions / 0 failed**; syntax-check **323 files**; `lag-mode` 50/50, `lag-member-checks` 8/8, `lag-member-down` 30/30, `l2` 28/28. Not done here: #1505's own test (`stated-range-is-the-enforced-range.test.js`) - the ceiling it asks for falls out of the shared check, but the four-point probe table is left to that issue.
A no channel-group undoes the group it names, and a port belongs to one bundle (#1510)
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 22s
CI / Test (Node 20) (pull_request) Successful in 3m7s
CI / Test (Node 18) (pull_request) Successful in 3m23s
CI / Server smoke test (pull_request) Successful in 33s
CI / Docker build (pull_request) Successful in 43s
CI / Test (Node 22) (pull_request) Successful in 3m7s
24ae336719
`no channel-group 9999` removed the port from whatever group it was really in: the dispatcher
dropped the id, so `no channel-group <anything>` acted on group 2 and said so. A technician
correcting a typo took the port out of the bundle that was carrying traffic.

- the id is read, range-checked (1-4096, both ends, shared with the set form), and an id that is
  not this port's group is refused naming both groups: `% Fa0/1 is not in channel group 5: it is
  in group 2`. A bare `no channel-group` keeps its meaning - whatever the port is in.
- a second `channel-group` moves the port instead of adding it to another group, and says
  `Interface Fa0/1 moved from channel group 2 to 10 (mode on)`.
- `show etherchannel summary` orders bundles by number: 2, 10, 100 - not 10, 100, 2.

test/no-command-touches-only-what-it-names.test.js: 4 new checks (11/11 in the file, 3 failing
before the handler changed).
jaylouisw scheduled this pull request to auto merge when all checks succeed 2026-11-15 21:03:05 +00:00
jaylouisw scheduled this pull request to auto merge when all checks succeed 2026-11-15 21:05:30 +00:00
jaylouisw force-pushed fix/1115-2035 from 24ae336719
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 22s
CI / Test (Node 20) (pull_request) Successful in 3m7s
CI / Test (Node 18) (pull_request) Successful in 3m23s
CI / Server smoke test (pull_request) Successful in 33s
CI / Docker build (pull_request) Successful in 43s
CI / Test (Node 22) (pull_request) Successful in 3m7s
to b62ed97347
All checks were successful
CI / Classify the change (pull_request) Successful in 14s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 18s
CI / Test (Node 18) (pull_request) Successful in 3m26s
CI / Test (Node 20) (pull_request) Successful in 3m12s
CI / Test (Node 22) (pull_request) Successful in 3m1s
CI / Server smoke test (pull_request) Successful in 29s
CI / Docker build (pull_request) Successful in 1m0s
2026-11-15 21:05:35 +00:00
Compare
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!1543
No description provided.