test(agent): fail hung/slow tests loudly (pytest-timeout + aws_session barrier bound)

agent/pyproject.toml

@@ -89,6 +89,7 @@ dev = [
     "pytest",
     "pygments==2.20.0",
     "pytest-cov==7.1.0",
+    "pytest-timeout==2.4.0",  # per-test wall-clock cap: a single hung test (network/subprocess/Bedrock without its own timeout) must fail LOUDLY with a traceback, not silently burn the whole build-verify budget (ABCA-684/686: one hang stalled the baseline build past its 3600s ceiling)
     "vulture==2.16",  # dead-code detection (#282): unused functions/classes ruff F can't see
 ]
 
@@ -130,6 +131,18 @@ ignore = [
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 pythonpath = ["src"]
+# pytest-timeout: hard wall-clock cap PER TEST. A unit test that blocks (an
+# unmocked network/subprocess/Bedrock call with no timeout of its own) otherwise
+# hangs the whole `mise run build` until the 3600s build-verify ceiling kills it,
+# turning one flaky test into an un-diagnosable 60-min stall (ABCA-684/686). With
+# this, the offending test fails with a dumped stack in 120s and the suite moves on.
+# method="signal" (SIGALRM) actually ABORTS the hung test — pytest runs
+# single-threaded in the main thread here, so the alarm interrupts even a blocked
+# C-level/socket call. method="thread" only PRINTS the stack and cannot interrupt a
+# syscall-blocked test (proven live on ABCA-688: the thread method dumped the stack
+# but the build kept hanging ~55 min until the 3600s ceiling). signal is the real fix.
+timeout = 120
+timeout_method = "signal"
 
 [tool.coverage.run]
 branch = true

agent/tests/test_aws_session.py

@@ -131,7 +131,15 @@ def _slow_build(_arn: str) -> Any:
             return MagicMock(name="scoped")
 
         def _worker() -> None:
-            start.wait()
+            # Bounded barrier wait. A bare Barrier(20).wait() blocks FOREVER if
+            # even one of the 20 threads never arrives (e.g. a worker reaped under
+            # container memory pressure, or thread creation throttled) — every
+            # survivor then hangs here and the main thread hangs in join() below,
+            # stalling the whole `mise run build` until the 3600s ceiling. This is
+            # the ECS-only flaky hang chased across ABCA-684/686/688 (pytest-timeout
+            # only fixed the SYMPTOM; this Barrier is the ROOT cause). A timeout
+            # makes the barrier raise BrokenBarrierError so the test fails fast.
+            start.wait(timeout=30)
             session = get_session()
             with lock:
                 results.append(session)
@@ -140,8 +148,11 @@ def _worker() -> None:
             threads = [threading.Thread(target=_worker) for _ in range(20)]
             for t in threads:
                 t.start()
+            # Bounded joins for the same reason — a worker that died before
+            # appending must not hang the suite; a leftover live thread trips the
+            # assertion below (results != 20) rather than blocking indefinitely.
             for t in threads:
-                t.join()
+                t.join(timeout=60)
 
         mk_build.assert_called_once()
         assert len(results) == 20