3 Commits

Author SHA1 Message Date
Alexander Whitestone
64a8ffa329 fix: watchdog single-instance guard, stop killing worker processes
- Pidfile guard prevents duplicate watchdog instances
- Removed check_zombies — was killing legitimate git push/clone from
  active workers (caused mass exit 143 failures)
- Replaced with check_disk that cleans clone dirs >1hr old
- Disabled gemini loop check (no API key configured)
- Workers use fresh clone per issue (no shared worktree contention)
- Simplified cleanup to rm -rf (no git worktree bookkeeping)
- Tested file_issue end-to-end — confirmed working

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-22 22:03:58 -04:00
Alexander Whitestone
431cddf7f2 fix: replace worktrees with fresh clones, fix watchdog issue filing
- Workers now git clone --depth=1 per issue instead of sharing base repos
  with worktrees. Eliminates all contention and branch collision errors.
- Watchdog file_issue uses temp file + sys.argv for safe JSON escaping
- Watchdog zombie detection only kills processes >5min old (was killing
  legitimate git pushes from active workers)
- Simplified cleanup to plain rm -rf (no worktree bookkeeping needed)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-22 21:35:56 -04:00
Alexander Whitestone
5f8129d346 feat: parallel workers for all agents, dynamic scaling, self-healing watchdog
- claude-loop: 7 workers default, scales up to 21, 5s cooldown
- gemini-loop: rewritten as parallel worker system (3→12), multi-repo,
  auto-clone, correct CLI flags (-p/--yolo), bash 3.2 compatible
- loop-watchdog: monitors all loops every 2min, auto-restarts dead loops,
  kills zombies, files Gitea issues for unfixable problems
- ops-helpers: added ops-wake-watchdog, ops-kill-watchdog
- All scripts use file-based PID tracking (bash 3.2 safe)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-22 19:22:18 -04:00