emrg: daemon single-instance admission + truthful old-pid wait (rant 2026-08-18T12:49:09) by argszero · Pull Request #841 · argszero/emrg

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>

argszero · 2026-08-18T06:00:37Z

Windows multi-client daemon coexistence (observed 2026-08-18): with GUI + TUI resident, stale-restart storms left 4 python -m emrg.server processes on different ports and kept killing in-flight evolution cycles (a cycle running 399+ tool-loop messages died at 11:14:15 to a stale-pid restart, then restarted from the same half-finished worktree state — never completing).

Fix 1 — daemon-side single-instance admission (emrg/server/daemon.py)

Before the pid-file check, probe the port file: if a live daemon already answers (is_server_running_sync), log and refuse to start a duplicate.
Rationale: stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid-file check alone then lets a second instance bind a new port. A port-liveness probe is the definitive signal.

Fix 2 — client-side truthful old-pid wait (emrg/client/daemon_manager.py)

The old daemon must be TRULY dead before the port file is removed and a new one spawns. cleanup_server() previously deleted the port file BEFORE the wait, so is_running() (a port-file probe) returned False instantly → new daemon spawned while the old one was still shutting down.
Now: SIGTERM → wait on the old PID itself (POSIX os.kill(pid, 0) probe; Windows keeps the port probe — os.kill(pid, 0) would TerminateProcess there) → SIGKILL fallback after 10s → cleanup_server() only after death is confirmed.

Tests (4 new)

test_serve_refuses_duplicate_when_daemon_alive — admission positive path
test_serve_proceeds_when_no_live_daemon — admission negative path
test_restart_waits_until_old_pid_dead_before_cleanup — wait ordering
test_restart_force_kills_stuck_old_pid — SIGKILL fallback

Local: pytest 936 passed + 1 skipped (937 collected), GUI 257/257, import + CLI green. Agent.md pytest count 933→937.

…2026-08-18T12:49:09) Windows multi-client (GUI + TUI, possibly different installs) stale-restart storms leave multiple emrg.server processes on different ports, interrupting in-flight evolution cycles for hours. Two fixes: 1. Server admission (daemon.py): before the pid-file check, probe the port file — if a live daemon already answers, refuse to start a duplicate. Stale-restart sequences can leave the pid file missing while an old daemon is still alive; the pid check alone then lets a second instance bind a new port (observed: 4 coexisting processes). 2. Client wait (daemon_manager.py): the old daemon must be TRULY dead before the port file is removed. cleanup_server() previously deleted the port file first, so is_running() (a port-file probe) returned False instantly and a new daemon spawned while the old one was still shutting down. Now waits on the old PID (POSIX os.kill(pid,0) probe; Windows keeps the port probe — os.kill(pid,0) would TerminateProcess there), SIGKILL fallback, then cleanup_server() only after death. Tests: admission positive + negative (serve refuses duplicate / proceeds when no live daemon), restart waits-for-death ordering, SIGKILL fallback on stuck pid. Agent.md pytest count 933→937.

argszero

✅ LGTM — cycle 709 (1/3). CI test + test-windows PASS (32105124982). Local: pytest 936+1 (937 collected), GUI 257/257, import + CLI green. Admission probe + truthful old-pid wait verified with 4 new tests (both daemon.py and daemon_manager.py paths).

argszero

✅ LGTM — cycle (2/3, fresh instance). Head 0c55c70, CI test + test-windows PASS (32105124982). Re-verified admission probe + truthful old-pid wait (cleanup_server after death confirmed).

…ge + #841's 4 tests)

argszero

✅ LGTM — cycle 711 (3rd). Head 96ca181 (conflict-resolved), CI test + test-windows PASS (32106785798), MERGEABLE/CLEAN. Three consecutive LGTMs (cycles 709/710/711), no ❌ — merging. This lands the Windows multi-instance daemon fix (rants 12:49:09 + 14:01:10).

v0.2.47 -> v0.2.48 across 8 files / 9 refs (standard release bump): - emrg/__init__.py, pyproject.toml, uv.lock (emrg entry) - emrg/gui/package.json + package-lock.json (x2) - packaging/build-runtime.sh, make-installer.sh, make-run-installer.sh Covers 4 commits since v0.2.47: #839 (TUI exit clear-screen), #840 (GUI countdown refresh), #841 (daemon single-instance admission), #842 (GUI task last-run meta). Rant 2026-08-18T14:46:47. Co-authored-by: EMRG Evolution <emrg@argszero.dev>