ltms - Gitea

ltms closed issue fleet/fleetd#211

2026-09-01 09:13:03 +02:00

A backend-exhausted pane with no ⏺ marker is classified as an empty scrape, so the credential is never quarantined

ltms commented on issue fleet/fleetd#211

2026-09-01 09:12:57 +02:00

A backend-exhausted pane with no ⏺ marker is classified as an empty scrape, so the credential is never quarantined

Merged to main as 049e7d9, with a follow-up at cabcd87.

The follow-up closed a real gap in the merged fix: the raw-scrape fallback's backend-error branch put only the matched line into the…

ltms closed issue fleet/fleetd#214

2026-09-01 09:12:48 +02:00

A plain claude-code spawn mints no session id, so the most common member on the fleet can never be resumed

ltms commented on issue fleet/fleetd#214

2026-09-01 09:12:42 +02:00

A plain claude-code spawn mints no session id, so the most common member on the fleet can never be resumed

Merged to main as bf0e09b and now proven live: fleet_list reports agentSessionId: c73c8bbf-57f4-4223-b6f3-99a33ab71dce for a plain fleet_spawn that passed neither sessionName nor…

ltms closed issue fleet/fleetd#220

2026-09-01 09:12:28 +02:00

Launch command is typed into the pane and silently cut at 1024 bytes

ltms opened issue fleet/fleetd#220

2026-09-01 09:12:17 +02:00

Launch command is typed into the pane and silently cut at 1024 bytes

ltms pushed to main at fleet/fleetd

2026-09-01 09:11:50 +02:00

c3fa1136d4 #220: keep the launch command inside the pane's 1024-byte line

ltms commented on pull request fleet/fleetd#218

2026-09-01 08:28:13 +02:00

#214: mint a session id on every claude-code spawn, so every member is resumable

Merged to main as bf0e09b. Full suite on the merged result: Tests run: 1079, Failures: 0, Errors: 0, Skipped: 0 — mvn clean install, unpiped, run by me.

You did the one thing the brief…

ltms opened issue fleet/fleetd#219

2026-09-01 08:27:55 +02:00

OpenCodeLauncher decides a member's config and discovery roots from fleetd's own filesystem — the #213 defect, twice

ltms commented on pull request fleet/fleetd#217

2026-09-01 08:27:14 +02:00

#213: fix ZDOTDIR credential scrub gate + directory under memberHerdrSocket

Merged to main as e1eb50c. Full suite on the merged result: Tests run: 1079, Failures: 0, Errors: 0, Skipped: 0 — mvn clean install, unpiped, run by me.

Both halves are fixed the way…

ltms commented on pull request fleet/fleetd#216

2026-09-01 08:26:54 +02:00

fleetd#211: classify BACKEND_EXHAUSTED/BACKEND_ERROR from the raw scrape as a fallback

Merged to main as 049e7d9. Full suite on the merged result: Tests run: 1079, Failures: 0, Errors: 0, Skipped: 0 — mvn clean install, unpiped, run by me.

The fix is right and the…

ltms pushed to main at fleet/fleetd

2026-09-01 08:26:25 +02:00

cabcd87b66 #211 follow-up: the raw-scrape fallback must carry the pane, not just the matched line

bf0e09b1a2 #214: mint a session id on every claude-code spawn, so every member is resumable

e1eb50ce65 #213: fix ZDOTDIR credential scrub gate + directory under memberHerdrSocket

049e7d9d54 fleetd#211: classify BACKEND_EXHAUSTED/BACKEND_ERROR from the raw scrape as a fallback

ff3b49cd1e #214: mint a session id on every claude-code spawn, so every member is resumable

Compare 7 commits »

ltms closed issue fleet/fleetd#209

2026-08-31 18:29:04 +02:00

agentSessionId is resolved once at spawn and frozen, so an opencode member never reports one

ltms commented on issue fleet/fleetd#209

2026-08-31 18:28:56 +02:00

agentSessionId is resolved once at spawn and frozen, so an opencode member never reports one

Proven live on the running daemon (pid 79448, jar ed3eab5c8baa, redeployed from 966c58a today).

I spawned a fresh opencode member on a brand-new worktree, so any id reported had to be…

ltms commented on issue fleet/fleetd#137

2026-08-31 18:24:05 +02:00

CB-629: after a bridge_ask round-trip, the worker's final reply orphans its ticket, which then reports a false failure

Fixed on main in two parts:

388aba7 — after answer()'s own bounded wait times out, the worker's later fleet_reply completes its own async ticket instead of leaving it PENDING until…

ltms closed pull request fleet/fleetd#215

2026-08-31 18:23:49 +02:00

Supersedes #205, fixes #137

ltms commented on pull request fleet/fleetd#215

2026-08-31 18:23:40 +02:00

Supersedes #205, fixes #137

Merged to main as 966c58a. Full suite on the merge result: Tests run: 1071, Failures: 0, Errors: 0, Skipped: 0 — mvn clean install, unpiped, run by me.

REPORT-cb137.md was left out…

ltms pushed to main at fleet/fleetd

2026-08-31 18:23:18 +02:00

966c58a3b8 #137: don't hand one stranded reply to every open async ticket (#215)

de026b8f8a #137 follow-up: document defect-1 and defect-2 reachability, drop unconstructible test

97f6c33a45 #137: don't guess when a target has more than one open async task

7d4a4339c2 Merge branch 'worker/cb-137-ask-ticket-e7760c-2' of https://git.ltms.dev/fleet/fleetd into worker/cb137-ambiguous-task-4df3d8-4

ee5f8b932b #137: complete an async ticket's own reply after answer() times out

Compare 5 commits »

ltms closed pull request fleet/fleetd#205

2026-08-31 18:02:04 +02:00

#137: complete an async ticket's reply after answer() times out

ltms commented on pull request fleet/fleetd#205

2026-08-31 18:01:43 +02:00

#137: complete an async ticket's reply after answer() times out

Closing: this work is already on main as 388aba7 ("#137: an answered turn's reply completes its own ticket, instead of a false failure"). I checked the content, not just the title — main…