From 0ea0c53e892ca4cb4ecc67eb7cb8dd11d6a8c8cd Mon Sep 17 00:00:00 2001
From: teknium1 <127238744+teknium1@users.noreply.github.com>
Date: Mon, 14 Sep 2026 21:16:05 -0700
Subject: [PATCH] fix: record a known exit status when the reader's wait()
 raises

The early-EOF reaper made _finish_reader return without publishing when
wait() raises, so the session stays tracked for later reconciliation.
That is right for the pipe path (_reconcile_local_exit can still reap via
session.process), but PTY sessions have no session.process: when
ptyprocess.wait raises (waitpid ECHILD after isalive() already reaped the
child) the exitstatus is known, yet poll() reported "running" forever.

Only leave the session tracked when exit_code() is still None; otherwise
record the known status and finish as before.

Review finding: PTY session whose pty.wait raises stays in _running forever (fail-open regression vs main)
---
 tests/tools/test_process_registry.py | 20 ++++++++++++++++++++
 tools/process_registry.py            |  8 ++++++--
 2 files changed, 26 insertions(+), 2 deletions(-)

diff --git a/tests/tools/test_process_registry.py b/tests/tools/test_process_registry.py
index 9bda3a1656..d057a36c1e 100644
--- a/tests/tools/test_process_registry.py
+++ b/tests/tools/test_process_registry.py
@@ -361,6 +361,26 @@ def test_failed_reader_wait_does_not_publish_false_completion(registry, monkeypa
     assert moved == []
 
 
+def test_failed_reader_wait_still_records_known_exit_status(registry, monkeypatch):
+    """A PTY child reaped by isalive() has its status; a raising wait must not lose it."""
+    session = _make_session(sid="proc_pty_wait_failed")
+    moved = []
+    monkeypatch.setattr(registry, "_move_to_finished", lambda _s: moved.append(_s.id))
+
+    registry._finish_reader(
+        session,
+        MagicMock(decode=MagicMock(return_value="")),
+        lambda _text: None,
+        "PTY",
+        MagicMock(side_effect=OSError("waitpid ECHILD")),
+        lambda: 9,
+    )
+
+    assert session.exited is True
+    assert session.exit_code == 9
+    assert moved == [session.id]
+
+
 # =========================================================================
 # Incremental UTF-8 decoding across chunk boundaries
 # (ported from openclaw/openclaw#112325)
diff --git a/tools/process_registry.py b/tools/process_registry.py
index 3eb2c4d3f4..8330437b45 100644
--- a/tools/process_registry.py
+++ b/tools/process_registry.py
@@ -1178,8 +1178,12 @@ class ProcessRegistry(ProcessCheckpointMixin):
         try:
             wait()
         except Exception as e:
-            logger.warning("%s wait failed; leaving process tracked: %s", label, e)
-            return
+            # A PTY child reaped by isalive() already has its exitstatus; only an
+            # unknown status must stay tracked for later reconciliation.
+            if exit_code() is None:
+                logger.warning("%s wait failed; leaving process tracked: %s", label, e)
+                return
+            logger.warning("%s wait failed; recording known exit status: %s", label, e)
         self._finish_exited(session, exit_code())
 
     @staticmethod