ltms - Gitea

ltms commented on pull request fleet/fleetd#236

2026-09-03 05:57:20 +02:00

fleetd #234: key opencode model check on resolved session id; make spawn-time quarantine actually happen

Merged to main in 838a701. Round 4 is the one that ships.

I verified the inversion with javac against the real interface file, not against a copy:

Probe.java:4: error: incompatible…

ltms commented on pull request fleet/fleetd#240

2026-09-03 05:57:06 +02:00

fleetd#201 Unit 1: typed backend-error classification

Merged to main in 26bafe8.

Verified by the lead on the integrated main (all of Units 1-4 plus #234): 1163 tests, 0 failures, 0 compile errors, BUILD SUCCESS.

Three mutations of production…

ltms pushed to main at fleet/fleetd

2026-09-03 05:56:49 +02:00

26bafe824b fleetd #201/#227 unit 1: classify a backend error at the scrape

838a701109 fleetd #234: carry the profile hint to the exhaustion sink

e5eb3534c7 fleetd #201/#227 unit 3: lead outage nudge

959c83534f fleetd #201/#227 unit 2: credential outage policy

31b028e860 fleetd #234 round 4: invert ExhaustionSink's abstract method so the bug class is unrepresentable

Compare 13 commits »

ltms closed pull request fleet/fleetd#237

2026-09-03 05:45:12 +02:00

CB-201: retain backend failure outcome

ltms pushed to main at fleet/fleetd

2026-09-03 05:45:08 +02:00

c3672f5472 fleetd #201/#227 unit 4: durable BACKEND_ERROR member outcome

bbf68f3e3c CB-201: cover losing completion CAS

fe2e5ede34 CB-201: retain backend failure outcome

Compare 3 commits »

ltms commented on pull request fleet/fleetd#237

2026-09-03 05:44:57 +02:00

CB-201: retain backend failure outcome

Verified by the lead and merged

Both rounds check out. I re-ran every claim myself rather than reading the report.

Round 1

Full mvn clean install from inside fleetd/ on fe2e5ed,…

ltms opened issue fleet/fleetd#241

2026-09-03 05:42:40 +02:00

Completion fallback can hand the lead back its own brief as the member's "report"

ltms commented on issue fleet/fleetd#234

2026-09-03 05:23:03 +02:00

#175's model read-back fires a FALSE POSITIVE on any member spawned without a worktree — and the quarantine it announces never happens

Review of PR #236 — defect 1 accepted, defect 2's fix is dead on the production path

Defect 1 (the model check keyed on directory instead of the resolved session id) is fixed correctly.…

ltms commented on issue fleet/fleetd#227

2026-09-03 05:22:39 +02:00

A backend outage is invisible at fleet level: repeated backend errors are never correlated, never cool off a credential, and never reach the lead

Refined together with #201 — see #201 for the full plan

An architect member refined #201 and this issue as one delivery program in five units. The design is on main: `docs/CB-201-227-Refin…

ltms commented on issue fleet/fleetd#201

2026-09-03 05:22:16 +02:00

Backend-error classification needs a real mechanism, not hard-coded strings (#164 points 3 and 4)

Refined into five units — design is on main

An architect member read this issue, #227, and the named source, and split the work. Full design: docs/CB-201-227-Refinement.md (merged to main…

ltms pushed to main at fleet/fleetd

2026-09-03 05:21:47 +02:00

7662e2d0c8 fleetd #201/#227: refine backend outage work

ltms merged pull request fleet/fleetd#235

2026-09-03 05:02:27 +02:00

CB-612: suppress unneeded token warnings

ltms pushed to main at fleet/fleetd

2026-09-03 05:02:27 +02:00

1c051c4e47 fleetd #115: a profile that needs no token no longer warns, and no longer crashes

adb7a67880 CB-612: support tokenless Claude Code profiles

723fe494e9 CB-612: suppress unneeded token warnings

Compare 3 commits »

ltms deleted branch worker/cb115-startup-warning-85c539-2 from fleet/fleetd

2026-09-03 05:02:27 +02:00

ltms opened issue fleet/fleetd#234

2026-09-03 04:26:09 +02:00

#175's model read-back fires a FALSE POSITIVE on any member spawned without a worktree — and the quarantine it announces never happens

ltms pushed to main at fleet/fleetd

2026-09-03 04:20:24 +02:00

0f08b93659 fleetd #176: spawn gate fails fast on a dead backend, and refines UNKNOWN safely

432c1d92d1 fleetd #176 review: cover the namePrefix guard, the adapter-refinement corroboration that had no test

60b7e67b42 fleetd #176: fail fast when the spawn gate's backend exits mid-wait, and safely refine its UNKNOWN status

Compare 3 commits »

ltms deleted branch worker/cb176-spawn-gate-3d75c8-2 from fleet/fleetd

2026-09-03 04:20:24 +02:00

ltms merged pull request fleet/fleetd#233

2026-09-03 04:20:23 +02:00

fleetd #176: fail-fast + safe UNKNOWN refinement in the spawn-readiness gate

ltms closed issue fleet/fleetd#232

2026-09-03 04:10:23 +02:00

autoCompactWindow is set by flag and never read back — the #175 shape in a second place

ltms commented on issue fleet/fleetd#232

2026-09-03 04:10:17 +02:00

autoCompactWindow is set by flag and never read back — the #175 shape in a second place

Checked. opencode does not report it. Closing with no code change.

This is outcome 3 from the ticket's own "What to do": "We checked and the backend does not report it." No probe was…