etherchannel: a group too wide for its new mode is refused, not converted (#2033) #2069

Merged
jaylouisw merged 1 commit from fix/1119-0202 into main 2026-11-19 02:49:17 +00:00
Owner

Refs #2033 — this closes the engine half that issue recorded and left open, not its docs transcript.

What was wrong. The EtherChannel cap was checked only where a port joins. A mode change is the one line that rewrites the cap a group already holds under, and it was unguarded: nine members fit one LACP group (cap sixteen), channel-group 1 mode on typed on a member moved the group to mode on — cap eight — and left all nine in it. Three lies followed from that one state:

  • show etherchannel summary printed nine members of a static bundle with Po1(SU), the ninth as Gi1/0/9(s) stand-alone yet still a member;
  • the next channel-group 1 mode on join was refused with "channel group 1 already holds 8 members" over a group holding nine — a number the operator cannot find anywhere;
  • the joining port was told Gi1/0/9 is hot-standby, in a bundle whose own key line says H - Hot-standby (LACP only).

Changed

  • engine/cli.js (_channelGroup): a mode change that would leave the group holding more members than the new mode allows is refused, nothing changed, naming the count, the cap, and the form that keeps the bundle (channel-group 1 mode active). The join refusal now names the group's real count, and when the line also re-modes, says what the count would become. The hot-standby receipt is only printed for a negotiating bundle.
  • docs/CLI.md: the paragraph that said the cap "is checked when a port joins" now says what is checked where, with the refusal transcript.
  • test/etherchannel-member-limit.test.js: 14 new assertions (refusal and its wording, group untouched, table still LACP, all nine members present, the tenth-join count, and an eight-member group that still converts so the guard is about the overflow rather than about conversion).

Gates. node scripts/syntax-check.js — 433 files OK. Full npm test — 356 suites, 12294 assertions, 0 failed.

Not done. The mode conversion of a group at the cap is still allowed in the other direction (mode on → mode active) — the cap grows there. Nothing in this console lets a static bundle hold nine, so a saved config carrying one is now the only way to reach the state; the saver does not write one.

Refs #2033 — this closes the **engine** half that issue recorded and left open, not its docs transcript. **What was wrong.** The EtherChannel cap was checked only where a port joins. A mode change is the one line that rewrites the cap a group already holds under, and it was unguarded: nine members fit one LACP group (cap sixteen), `channel-group 1 mode on` typed on a member moved the group to `mode on` — cap eight — and left all nine in it. Three lies followed from that one state: - `show etherchannel summary` printed nine members of a static bundle with `Po1(SU)`, the ninth as `Gi1/0/9(s)` stand-alone yet still a member; - the next `channel-group 1 mode on` join was refused with "channel group 1 already holds 8 members" over a group holding nine — a number the operator cannot find anywhere; - the joining port was told `Gi1/0/9 is hot-standby`, in a bundle whose own key line says `H - Hot-standby (LACP only)`. **Changed** - `engine/cli.js` (`_channelGroup`): a mode change that would leave the group holding more members than the new mode allows is refused, nothing changed, naming the count, the cap, and the form that keeps the bundle (`channel-group 1 mode active`). The join refusal now names the group's real count, and when the line also re-modes, says what the count would become. The hot-standby receipt is only printed for a negotiating bundle. - `docs/CLI.md`: the paragraph that said the cap "is checked when a port joins" now says what is checked where, with the refusal transcript. - `test/etherchannel-member-limit.test.js`: 14 new assertions (refusal and its wording, group untouched, table still LACP, all nine members present, the tenth-join count, and an eight-member group that still converts so the guard is about the overflow rather than about conversion). **Gates.** `node scripts/syntax-check.js` — 433 files OK. Full `npm test` — 356 suites, 12294 assertions, 0 failed. **Not done.** The mode conversion of a group *at* the cap is still allowed in the other direction (`mode on` → `mode active`) — the cap grows there. Nothing in this console lets a static bundle hold nine, so a saved config carrying one is now the only way to reach the state; the saver does not write one.
etherchannel: a group too wide for its new mode is refused, not converted (#2033)
All checks were successful
CI / Classify the change (pull_request) Successful in 7s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 10s
CI / Server smoke test (pull_request) Successful in 10s
CI / Docker build (pull_request) Successful in 22s
CI / Test (Node 20) (pull_request) Successful in 57s
CI / Test (Node 18) (pull_request) Successful in 1m1s
CI / Test (Node 22) (pull_request) Successful in 57s
CI / Arm auto-merge (pull_request) Successful in 4s
ee7aa2a804
forgejo-actions force-pushed fix/1119-0202 from ee7aa2a804
All checks were successful
CI / Classify the change (pull_request) Successful in 7s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 10s
CI / Server smoke test (pull_request) Successful in 10s
CI / Docker build (pull_request) Successful in 22s
CI / Test (Node 20) (pull_request) Successful in 57s
CI / Test (Node 18) (pull_request) Successful in 1m1s
CI / Test (Node 22) (pull_request) Successful in 57s
CI / Arm auto-merge (pull_request) Successful in 4s
to f4b0186a98 2026-11-19 02:11:29 +00:00
Compare
jaylouisw force-pushed fix/1119-0202 from f4b0186a98 to dad118ffd4
All checks were successful
CI / Classify the change (pull_request) Successful in 7s
PR rebase / Rebase onto main and arm auto-merge (pull_request) Successful in 5s
CI / Server smoke test (pull_request) Successful in 9s
CI / Docker build (pull_request) Successful in 22s
CI / Test (Node 20) (pull_request) Successful in 53s
CI / Test (Node 22) (pull_request) Successful in 56s
CI / Test (Node 18) (pull_request) Successful in 59s
CI / Arm auto-merge (pull_request) Successful in 3s
2026-11-19 02:47:13 +00:00
Compare
Sign in to join this conversation.
No reviewers
No milestone
No project
No assignees
2 participants
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!2069
No description provided.