ltms - Gitea

ltms closed issue fleet/fleetd#437

2026-09-10 12:21:13 +02:00

fleet_ack says "acknowledged <msgId>" for a held peer message it never touches

ltms commented on issue fleet/fleetd#437

2026-09-10 12:21:07 +02:00

fleet_ack says "acknowledged <msgId>" for a held peer message it never touches

Closed by PR #448.

What shipped

ReplyInbox.ack and MessageService.ackReply return boolean instead of void. false means nothing was removed and is not an error. FleetMcp.ack now…

ltms pushed to main at fleet/fleetd

2026-09-10 12:20:50 +02:00

c11ad71ed0 Merge #448: fleet_ack errors instead of claiming success on a miss (fleetd #437)

5289eb509f fleetd #437: pin the ack hit/miss contract in the AMQP contract test

703a05db41 fleetd #437: fleet_ack errors instead of claiming success on a miss

Compare 3 commits »

ltms merged pull request fleet/fleetd#448

2026-09-10 12:20:49 +02:00

fleetd #437: fleet_ack errors instead of claiming success on a miss

ltms opened issue fleet/fleetd#453

2026-09-10 12:18:15 +02:00

PeerLauncher.place(role) and defaultProfileFor(role) are default methods that silently ignore the role

ltms closed issue fleet/fleetd#450

2026-09-10 12:17:12 +02:00

PeerLauncher's default spawn(req, decision) is still the #425 defect, and nothing tests it — make it abstract

ltms commented on issue fleet/fleetd#450

2026-09-10 12:17:06 +02:00

PeerLauncher's default spawn(req, decision) is still the #425 defect, and nothing tests it — make it abstract

Closed by PR #451.

What landed

PeerLauncher.spawn(SpawnRequest, PlacementDecision) is abstract. HerdrPeerLauncher — the abstract base of ClaudeCodeLauncher and OpenCodeLauncher —…

ltms pushed to main at fleet/fleetd

2026-09-10 12:16:49 +02:00

bdcf285265 Merge #451: make PeerLauncher.spawn(req, decision) abstract (fleetd #450)

cfebc575ea fleetd #450: make PeerLauncher.spawn(SpawnRequest, PlacementDecision) abstract

Compare 2 commits »

ltms merged pull request fleet/fleetd#451

2026-09-10 12:16:47 +02:00

fleetd #450: make PeerLauncher.spawn(SpawnRequest, PlacementDecision) abstract

ltms commented on issue fleet/fleetd#449

2026-09-10 12:07:02 +02:00

The herdr contract tests were left behind by the protocol-19 port, and nothing runs them

Measured the whole group again on main at 822327e, and it changes the CI part of this ticket. Delegated with these numbers.

The group's actual result

`mvn -B -Pcontract test -Dgroups=contr…

ltms commented on issue fleet/fleetd#444

2026-09-10 12:03:02 +02:00

PlacementDecision's javadoc claims the place()-to-spawn() window is closed, but no test pins it — and PeerLauncher's default spawn(req, decision) reopens it

One honest note on closing this. Criterion 3 asked to either keep the default and test it, or make it abstract. #447 kept it and explained why, which is half of option 1. The test half was not…

ltms opened issue fleet/fleetd#450

2026-09-10 12:02:44 +02:00

PeerLauncher's default spawn(req, decision) is still the #425 defect, and nothing tests it — make it abstract

ltms closed issue fleet/fleetd#444

2026-09-10 12:01:36 +02:00

PlacementDecision's javadoc claims the place()-to-spawn() window is closed, but no test pins it — and PeerLauncher's default spawn(req, decision) reopens it

ltms commented on issue fleet/fleetd#444

2026-09-10 12:01:29 +02:00

PlacementDecision's javadoc claims the place()-to-spawn() window is closed, but no test pins it — and PeerLauncher's default spawn(req, decision) reopens it

Closed by PR #447, merged as 822327e. The merged tree is byte-identical to the branch tip I tested (git diff --stat origin/main e4c703a printed nothing).

What landed

One test, `CompositePe…

ltms pushed to main at fleet/fleetd

2026-09-10 12:01:00 +02:00

822327eed5 Merge #447: pin the place()-to-spawn() window PlacementDecision closes (fleetd #444)

e4c703a51a fleetd #444: separate the adapter's fallback default from the decided profile

3f036b2a62 fleetd #444: pin the place()-to-spawn() window PlacementDecision closes

Compare 3 commits »

ltms merged pull request fleet/fleetd#447

2026-09-10 12:00:59 +02:00

fleetd #444: test that pins the place()-to-spawn() window PlacementDecision closes

ltms commented on issue fleet/fleetd#437

2026-09-10 11:59:28 +02:00

fleet_ack says "acknowledged <msgId>" for a held peer message it never touches

Decision: option 2, refuse. Plus two corrections to my own text above.

The fix is in PR #448. Before that, two things in the body are wrong or too strong, and both change how the next reader…

ltms opened issue fleet/fleetd#449

2026-09-10 11:52:38 +02:00

The herdr contract tests were left behind by the protocol-19 port, and nothing runs them

ltms commented on issue fleet/fleetd#400

2026-09-10 11:49:26 +02:00

SECURITY (receipt): the scrub reports a name as blanked when eval returned 0 without blanking it

Already fixed and merged. Closing.

c670792 ("#400: classify the blanking loop's result on the observed value, not eval's exit status"), merged as 3d61af6. I confirmed it myself: `git…

ltms opened issue fleet/fleetd#446

2026-09-10 11:40:19 +02:00

The model gate can be turned off at runtime but its usage-limit detector cannot be turned on at runtime