Sixth and final diagnostic pass — after all 4
cascade fixes landed (FD hygiene, pidfd wait,
`_parent_chan_cs` wiring, bounded peer-clear), the
actual last gate on
`test_nested_multierrors[subint_forkserver]`
turned out to be **pytest's default
`--capture=fd` stdout/stderr capture**, not
anything in the runtime cascade.
Empirical result: `pytest -s` → test PASSES in
6.20s. Default `--capture=fd` → hangs forever.
Mechanism: pytest replaces the parent's fds 1,2
with pipe write-ends it reads from. Fork children
inherit those pipes (since `_close_inherited_fds`
correctly preserves stdio). The error-propagation
cascade in a multi-level cancel test generates
7+ actors each logging multiple `RemoteActorError`
/ `ExceptionGroup` tracebacks — enough output to
fill Linux's 64KB pipe buffer. Writes block,
subactors can't progress, processes don't exit,
`_ForkedProc.wait` hangs.
Self-critical aside: I earlier tested w/ and w/o
`-s` and both hung, concluding "capture-pipe
ruled out". That was wrong — at that time fixes
1-4 weren't all in place, so the test was
failing at deeper levels long before reaching
the "produce lots of output" phase. Once the
cascade could actually tear down cleanly, enough
output flowed to hit the pipe limit. Order-of-
operations mistake: ruling something out based
on a test that was failing for a different
reason.
Deats,
- `subint_forkserver_test_cancellation_leak_issue
.md`: new section "Update — VERY late: pytest
capture pipe IS the final gate" w/ DIAG timeline
showing `trio.run` fully returns, diagnosis of
pipe-fill mechanism, retrospective on the
earlier wrong ruling-out, and fix direction
(redirect subactor stdout/stderr to `/dev/null`
in fork-child prelude, conditional on
pytest-detection or opt-in flag)
- `tests/test_cancellation.py`: skip-mark reason
rewritten to describe the capture-pipe gate
specifically; cross-refs the new doc section
- `tests/spawn/test_subint_forkserver.py`: the
orphan-SIGINT test regresses back to xfail.
Previously passed after the FD-hygiene fix,
but the new `wait_for_no_more_peers(
move_on_after=3.0)` bound in `async_main`'s
teardown added up to 3s latency, pushing
orphan-subactor exit past the test's 10s poll
window. Real fix: faster orphan-side teardown
OR extend poll window to 15s
No runtime code changes in this commit — just
test-mark adjustments + doc wrap-up.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Fifth diagnostic pass pinpointed the hang to
`async_main`'s finally block — every stuck actor
reaches `FINALLY ENTER` but never `RETURNING`.
Specifically `await ipc_server.wait_for_no_more_
peers()` never returns when a peer-channel handler
is stuck: the `_no_more_peers` Event is set only
when `server._peers` empties, and stuck handlers
keep their channels registered.
Wrap the call in `trio.move_on_after(3.0)` + a
warning-log on timeout that records the still-
connected peer count. 3s is enough for any
graceful cancel-ack round-trip; beyond that we're
in bug territory and need to proceed with local
teardown so the parent's `_ForkedProc.wait()` can
unblock. Defensive-in-depth regardless of the
underlying bug — a local finally shouldn't block
on remote cooperation forever.
Verified: with this fix, ALL 15 actors reach
`async_main: RETURNING` (up from 10/15 before).
Test still hangs past 45s though — there's at
least one MORE unbounded wait downstream of
`async_main`. Candidates enumerated in the doc
update (`open_root_actor` finally /
`actor.cancel()` internals / trio.run bg tasks /
`_serve_ipc_eps` finally). Skip-mark stays on
`test_nested_multierrors[subint_forkserver]`.
Also updates
`subint_forkserver_test_cancellation_leak_issue.md`
with the new pinpoint + summary of the 6-item
investigation win list:
1. FD hygiene fix (`_close_inherited_fds`) —
orphan-SIGINT closed
2. pidfd-based `_ForkedProc.wait` — cancellable
3. `_parent_chan_cs` wiring — shielded parent-chan
loop now breakable
4. `wait_for_no_more_peers` bound — THIS commit
5. Ruled-out hypotheses: tree-kill missing, stuck
socket recv, capture-pipe fill (all wrong)
6. Remaining unknown: at least one more unbounded
wait in the teardown cascade above `async_main`
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Fourth diagnostic pass — instrument `_worker`'s
fork-child branch (`pre child_target()` / `child_
target RETURNED rc=N` / `about to os._exit(rc)`)
and `_trio_main` boundaries (`about to trio.run` /
`trio.run RETURNED NORMALLY` / `FINALLY`). Test
config: depth=1/breadth=2 = 1 root + 14 forked =
15 actors total.
Fresh-run results,
- **9 processes complete the full flow**:
`trio.run RETURNED NORMALLY` → `child_target
RETURNED rc=0` → `os._exit(0)`. These are tree
LEAVES (errorers) plus their direct parents
(depth-0 spawners) — they actually exit
- **5 processes stuck INSIDE `trio.run(trio_
main)`**: hit "about to trio.run" but never
see "trio.run RETURNED NORMALLY". These are
root + top-level spawners + one intermediate
The deadlock is in `async_main` itself, NOT the
peer-channel loops. Specifically, the outer
`async with root_tn:` in `async_main` never exits
for the 5 stuck actors, so the cascade wedges:
trio.run never returns
→ _trio_main finally never runs
→ _worker never reaches os._exit(rc)
→ process never dies
→ parent's _ForkedProc.wait() blocks
→ parent's nursery hangs
→ parent's async_main hangs
→ (recurse up)
The precise new question: **what task in the 5
stuck actors' `async_main` never completes?**
Candidates:
1. shielded parent-chan `process_messages` task
in `root_tn` — but we cancel it via
`_parent_chan_cs.cancel()` in `Actor.cancel()`,
which only runs during
`open_root_actor.__aexit__`, which itself runs
only after `async_main`'s outer unwind — which
doesn't happen. So the shield isn't broken in
this path.
2. `actor_nursery._join_procs.wait()` or similar
inline in the backend `*_proc` flow.
3. `_ForkedProc.wait()` on a grandchild that DID
exit — but pidfd_open watch didn't fire (race
between `pidfd_open` and the child exiting?).
Most specific next probe: add DIAG around
`_ForkedProc.wait()` enter/exit to see whether
pidfd-based wait returns for every grandchild
exit. If a stuck parent's `_ForkedProc.wait()`
never returns despite its child exiting → pidfd
mechanism has a race bug under nested forkserver.
Asymmetry observed in the cascade tree: some d=0
spawners exit cleanly, others stick, even though
they started identically. Not purely depth-
determined — some race condition in nursery
teardown when multiple siblings error
simultaneously.
No code changes — diagnosis-only.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Third diagnostic pass on
`test_nested_multierrors[subint_forkserver]` hang.
Two prior hypotheses ruled out + a new, more
specific deadlock shape identified.
Ruled out,
- **capture-pipe fill** (`-s` flag changes test):
retested explicitly — `test_nested_multierrors`
hangs identically with and without `-s`. The
earlier observation was likely a competing
pytest process I had running in another session
holding registry state
- **stuck peer-chan recv that cancel can't
break**: pivot from the prior pass. With
`handle_stream_from_peer` instrumented at ENTER
/ `except trio.Cancelled:` / finally: 40
ENTERs, ZERO `trio.Cancelled` hits. Cancel never
reaches those tasks at all — the recvs are
fine, nothing is telling them to stop
Actual deadlock shape: multi-level mutual wait.
root blocks on spawner.wait()
spawner blocks on grandchild.wait()
grandchild blocks on errorer.wait()
errorer Actor.cancel() ran, but proc
never exits
`Actor.cancel()` fired in 12 PIDs — but NOT in
root + 2 direct spawners. Those 3 have peer
handlers stuck because their own `Actor.cancel()`
never runs, which only runs when the enclosing
`tractor.open_nursery()` exits, which waits on
`_ForkedProc.wait()` for the child pidfd to
signal, which only signals when the child
process fully exits.
Refined question: **why does an errorer process
not exit after its `Actor.cancel()` completes?**
Three hypotheses (unverified):
1. `_parent_chan_cs.cancel()` fires but the
shielded loop's recv is stuck in a way cancel
still can't break
2. `async_main`'s post-cancel unwind has other
tasks in `root_tn` awaiting something that
never arrives (e.g. outbound IPC reply)
3. `os._exit(rc)` in `_worker` never runs because
`_child_target` never returns
Next-session probes (priority order):
1. instrument `_worker`'s fork-child branch —
confirm whether `child_target()` returns /
`os._exit(rc)` is reached for errorer PIDs
2. instrument `async_main`'s final unwind — see
which await in teardown doesn't complete
3. compare under `trio_proc` backend at the
equivalent level to spot divergence
No code changes — diagnosis-only.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Three places that previously swallowed exceptions silently now log via
`log.exception()` so they surface in the runtime log when something
weird happens — easier to track down sneaky failures in the
fork-from-worker-thread / subint-bootstrap primitives.
Deats,
- `_close_inherited_fds()`: post-fork child's per-fd `os.close()`
swallow now logs the fd that failed to close. The comment notes the
expected failure modes (already-closed-via-listdir-race,
otherwise-unclosable) — both still fine to ignore semantically, but
worth flagging in the log.
- `fork_from_worker_thread()` parent-side timeout branch: the
`os.close(rfd)` + `os.close(wfd)` cleanup now logs each pipe-fd close
failure separately before raising the `worker thread didn't return`
RuntimeError.
- `run_subint_in_worker_thread._drive()`: when
`_interpreters.exec(interp_id, bootstrap)` raises a `BaseException`,
log the full call signature (interp_id + bootstrap) along with the
captured exception, before stashing into `err` for the outer caller.
Behavior unchanged — only adds observability.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Two new sections in
`subint_forkserver_test_cancellation_leak_issue.md`
documenting continued investigation of the
`test_nested_multierrors[subint_forkserver]` peer-
channel-loop hang:
1. **"Attempted fix (DID NOT work) — hypothesis
(3)"**: tried sync-closing peer channels' raw
socket fds from `_serve_ipc_eps`'s finally block
(iterate `server._peers`, `_chan._transport.
stream.socket.close()`). Theory was that sync
close would propagate as `EBADF` /
`ClosedResourceError` into the stuck
`recv_some()` and unblock it. Result: identical
hang. Either trio holds an internal fd
reference that survives external close, or the
stuck recv isn't even the root blocker. Either
way: ruled out, experiment reverted, skip-mark
restored.
2. **"Aside: `-s` flag changes behavior for peer-
intensive tests"**: noticed
`test_context_stream_semantics.py` under
`subint_forkserver` hangs with default
`--capture=fd` but passes with `-s`
(`--capture=no`). Working hypothesis: subactors
inherit pytest's capture pipe (fds 1,2 — which
`_close_inherited_fds` deliberately preserves);
verbose subactor logging fills the buffer,
writes block, deadlock. Fix direction (if
confirmed): redirect subactor stdout/stderr to
`/dev/null` or a file in `_actor_child_main`.
Not a blocker on the main investigation;
deserves its own mini-tracker.
Both sections are diagnosis-only — no code changes
in this commit.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Two-part stopgap for the still-hanging
`test_nested_multierrors[subint_forkserver]`:
1. Skip-mark the test via
`@pytest.mark.skipon_spawn_backend('subint_forkserver',
reason=...)` so it stops blocking the test
matrix while the remaining bug is being chased.
The reason string cross-refs the conc-anal doc
for full context.
2. Update the conc-anal doc
(`subint_forkserver_test_cancellation_leak_issue.md`) with the
empirical state after the three nested- cancel fix commits
(`0cd0b633` FD scrub + `fe540d02` pidfd wait + `57935804` parent-chan
shield break) landed, narrowing the remaining hang from "everything
broken" to "peer-channel loops don't exit on `service_tn` cancel".
Deats from the DIAGDEBUG instrumentation pass,
- 80 `process_messages` ENTERs, 75 EXITs → 5 stuck
- ALL 40 `shield=True` ENTERs matched EXIT — the
`_parent_chan_cs.cancel()` wiring from `57935804`
works as intended for shielded loops.
- the 5 stuck loops are all `shield=False` peer-
channel handlers in `handle_stream_from_peer`
(inbound connections handled by
`stream_handler_tn`, which IS `service_tn` in the
current config).
- after `_parent_chan_cs.cancel()` fires, NEW
shielded loops appear on the session reg_addr
port — probably discovery-layer reconnection;
doesn't block teardown but indicates the cascade
has more moving parts than expected.
The remaining unknown: why don't the 5 peer-channel loops exit when
`service_tn.cancel_scope.cancel()` fires? They're not shielded, they're
inside the service_tn scope, a standard cancel should propagate through.
Some fork-config-specific divergence keeps them alive. Doc lists three
follow-up experiments (stackscope dump, side-by-side `trio_proc`
comparison, audit of the `tractor/ipc/_server.py:448` `except
trio.Cancelled:` path).
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Completes the nested-cancel deadlock fix started in
0cd0b633 (fork-child FD scrub) and fe540d02 (pidfd-
cancellable wait). The remaining piece: the parent-
channel `process_messages` loop runs under
`shield=True` (so normal cancel cascades don't kill
it prematurely), and relies on EOF arriving when the
parent closes the socket to exit naturally.
Under exec-spawn backends (`trio_proc`, mp) that EOF
arrival is reliable — parent's teardown closes the
handler-task socket deterministically. But fork-
based backends like `subint_forkserver` share enough
process-image state that EOF delivery becomes racy:
the loop parks waiting for an EOF that only arrives
after the parent finishes its own teardown, but the
parent is itself blocked on `os.waitpid()` for THIS
actor's exit. Mutual wait → deadlock.
Deats,
- `async_main` stashes the cancel-scope returned by
`root_tn.start(...)` for the parent-chan
`process_messages` task onto the actor as
`_parent_chan_cs`
- `Actor.cancel()`'s teardown path (after
`ipc_server.cancel()` + `wait_for_shutdown()`)
calls `self._parent_chan_cs.cancel()` to
explicitly break the shield — no more waiting for
EOF delivery, unwinding proceeds deterministically
regardless of backend
- inline comments on both sites explain the mutual-
wait deadlock + why the explicit cancel is
backend-agnostic rather than a forkserver-specific
workaround
With this + the prior two fixes, the
`subint_forkserver` nested-cancel cascade unwinds
cleanly end-to-end.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Two coordinated improvements to the `subint_forkserver` backend:
1. Replace `trio.to_thread.run_sync(os.waitpid, ...,
abandon_on_cancel=False)` in `_ForkedProc.wait()`
with `trio.lowlevel.wait_readable(pidfd)`. The
prior version blocked a trio cache thread on a
sync syscall — outer cancel scopes couldn't
unwedge it when something downstream got stuck.
Same pattern `trio.Process.wait()` and
`proc_waiter` (the mp backend) already use.
2. Drop the `@pytest.mark.xfail(strict=True)` from
`test_orphaned_subactor_sigint_cleanup_DRAFT` —
the test now PASSES after 0cd0b633 (fork-child
FD scrub). Same root cause as the nested-cancel
hang: inherited IPC/trio FDs were poisoning the
child's event loop. Closing them lets SIGINT
propagation work as designed.
Deats,
- `_ForkedProc.__init__` opens a pidfd via
`os.pidfd_open(pid)` (Linux 5.3+, Python 3.9+)
- `wait()` parks on `trio.lowlevel.wait_readable()`,
then non-blocking `waitpid(WNOHANG)` to collect
the exit status (correct since the pidfd signal
IS the child-exit notification)
- `ChildProcessError` swallow handles the rare race
where someone else reaps first
- pidfd closed after `wait()` completes (one-shot
semantics) + `__del__` belt-and-braces for
unexpected-teardown paths
- test docstring's `@xfail` block replaced with a
`# NOTE` comment explaining the historical
context + cross-ref to the conc-anal doc; test
remains in place as a regression guard
The two changes are interdependent — the
cancellable `wait()` matters for the same nested-
cancel scenarios the FD scrub fixes, since the
original deadlock had trio cache workers wedged in
`os.waitpid` swallowing the outer cancel.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Implements fix-direction (1)/blunt-close-all-FDs from
b71705bd (`subint_forkserver` nested-cancel hang
diag), targeting the multi-level cancel-cascade
deadlock in
`test_nested_multierrors[subint_forkserver]`.
The diagnosis doc voted for surgical FD cleanup via
`actor.ipc_server` handle as the cleanest approach,
but going blunt is actually the right call: after
`os.fork()`, the child immediately enters
`_actor_child_main()` which opens its OWN IPC
sockets / wakeup-fd / epoll-fd / etc. — none of the
parent's FDs are needed. Closing everything except
stdio is safe AND defends against future
listener/IPC additions to the parent inheriting
silently into children.
Deats,
- new `_close_inherited_fds(keep={0,1,2}) -> int`
helper. Linux fast-path enumerates `/proc/self/fd`;
POSIX fallback uses `RLIMIT_NOFILE` range. Matches
the stdlib `subprocess._posixsubprocess.close_fds`
strategy. Returns close-count for sanity logging
- wire into `fork_from_worker_thread._worker()`'s
post-fork child prelude — runs immediately after
the pid-pipe `os.close(rfd/wfd)`, before the user
`child_target` callable executes
- docstring cross-refs the diagnosis doc + spells
out the FD-inheritance-cascade mechanism and why
the close-all approach is safe for our spawn shape
Validation pending: re-run `test_nested_multierrors[subint_forkserver]`
to confirm the deadlock is gone.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Major rewrite of
`subint_forkserver_test_cancellation_leak_issue.md`
after empirical investigation revealed the earlier
"descendant-leak + missing tree-kill" diagnosis
conflated two unrelated symptoms:
1. **5-zombie leak holding `:1616`** — turned out to
be a self-inflicted cleanup bug: `pkill`-ing a bg
pytest task (SIGTERM/SIGKILL, no SIGINT) skipped
the SC graceful cancel cascade entirely. Codified
the real fix — SIGINT-first ladder w/ bounded
wait before SIGKILL — in e5e2afb5 (`run-tests`
SKILL) and
`feedback_sc_graceful_cancel_first.md`.
2. **`test_nested_multierrors[subint_forkserver]`
hangs indefinitely** — the actual backend bug,
and it's a deadlock not a leak.
Deats,
- new diagnosis: all 5 procs are kernel-`S` in
`do_epoll_wait`; pytest-main's trio-cache workers
are in `os.waitpid` waiting for children that are
themselves waiting on IPC that never arrives —
graceful `Portal.cancel_actor` cascade never
reaches its targets
- tree-structure evidence: asymmetric depth across
two identical `run_in_actor` calls — child 1
(3 threads) spawns both its grandchildren; child 2
(1 thread) never completes its first nursery
`run_in_actor`. Smells like a race on fork-
inherited state landing differently per spawn
ordering
- new hypothesis: `os.fork()` from a subactor
inherits the ROOT parent's IPC listener FDs
transitively. Grandchildren end up with three
overlapping FD sets (own + direct-parent + root),
so IPC routing becomes ambiguous. Predicts bug
scales with fork depth — matches reality: single-
level spawn works, multi-level hangs
- ruled out: `_ForkedProc.kill()` tree-kill (never
reaches hard-kill path), `:1616` contention (fixed
by `reg_addr` fixture wiring), GIL starvation
(each subactor has its own OS process+GIL),
child-side KBI absorption (`_trio_main` only
catches KBI at `trio.run()` callsite, reached
only on trio-loop exit)
- four fix directions ranked: (1) blanket post-fork
`closerange()`, (2) `FD_CLOEXEC` + audit,
(3) targeted FD cleanup via `actor.ipc_server`
handle, (4) `os.posix_spawn` w/ `file_actions`.
Vote: (3) — surgical, doesn't break the "no exec"
design of `subint_forkserver`
- standalone repro added (`spawn_and_error(breadth=
2, depth=1)` under `trio.fail_after(20)`)
- stopgap: skip `test_nested_multierrors` + multi-
level-spawn tests under the backend via
`@pytest.mark.skipon_spawn_backend(...)` until
fix lands
Killing the "tree-kill descendants" fix-direction
section: it addressed a bug that didn't exist.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
The previous cleanup recipe went straight to
SIGTERM+SIGKILL, which hides bugs: tractor is
structured concurrent — `_trio_main` catches SIGINT
as an OS-cancel and cascades `Portal.cancel_actor`
over IPC to every descendant. So a graceful SIGINT
exercises the actual SC teardown path; if it hangs,
that's a real bug to file (the forkserver `:1616`
zombie was originally suspected to be one of these
but turned out to be a teardown gap in
`_ForkedProc.kill()` instead).
Deats,
- step 1: `pkill -INT` scoped to `$(pwd)/py*` — no
sleep yet, just send the signal
- step 2: bounded wait loop (10 × 0.3s = ~3s) using
`pgrep` to poll for exit. Loop breaks early on
clean exit
- step 3: `pkill -9` only if graceful timed out, w/
a logged escalation msg so it's obvious when SC
teardown didn't complete
- step 4: same SIGINT-first ladder for the rare
`:1616`-holding zombie that doesn't match the
cmdline pattern (find PID via `ss -tlnp`, then
`kill -INT NNNN; sleep 1; kill -9 NNNN`)
- steps 5-6: UDS-socket `rm -f` + re-verify
unchanged
Goal: surface real teardown bugs through the test-
cleanup workflow instead of papering over them with
`-9`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Stopgap companion to d0121960 (`subint_forkserver`
test-cancellation leak doc): five tests in
`tests/test_cancellation.py` were running against the
default `:1616` registry, so any leaked
`subint-forkserv` descendant from a prior test holds
the port and blows up every subsequent run with
`TooSlowError` / "address in use". Thread the
session-unique `reg_addr` fixture through so each run
picks its own port — zombies can no longer poison
other tests (they'll only cross-contaminate whatever
happens to share their port, which is now nothing).
Deats,
- add `reg_addr: tuple` fixture param to:
- `test_cancel_infinite_streamer`
- `test_some_cancels_all`
- `test_nested_multierrors`
- `test_cancel_via_SIGINT`
- `test_cancel_via_SIGINT_other_task`
- explicitly pass `registry_addrs=[reg_addr]` to the
two `open_nursery()` calls that previously had no
kwargs at all (in `test_cancel_via_SIGINT` and
`test_cancel_via_SIGINT_other_task`)
- add bounded `@pytest.mark.timeout(7, method='thread')`
to `test_nested_multierrors` so a hung run doesn't
wedge the whole session
Still doesn't close the real leak — the
`subint_forkserver` backend's `_ForkedProc.kill()` is
PID-scoped not tree-scoped, so grandchildren survive
teardown regardless of registry port. This commit is
just blast-radius containment until that fix lands.
See `ai/conc-anal/
subint_forkserver_test_cancellation_leak_issue.md`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
New `ai/conc-anal/
subint_forkserver_test_cancellation_leak_issue.md`
captures a descendant-leak surfaced while wiring
`subint_forkserver` into the full test matrix:
running `tests/test_cancellation.py` under
`--spawn-backend=subint_forkserver` reproducibly
leaks **exactly 5** `subint-forkserv` comm-named
child processes that survive session exit, each
holding a `LISTEN` on `:1616` (the tractor default
registry addr) — and therefore poisons every
subsequent test session that defaults to that addr.
Deats,
- TL;DR + ruled-out checks confirming the procs are
ours (not piker / other tractor-embedding apps) —
`/proc/$pid/cmdline` + cwd both resolve to this
repo's `py314/` venv
- root cause: `_ForkedProc.kill()` is PID-scoped
(plain `os.kill(SIGKILL)` to the direct child),
not tree-scoped — grandchildren spawned during a
multi-level cancel test get reparented to init and
inherit the registry listen socket
- proposed fix directions ranked: (1) put each
forkserver-spawned subactor in its own process-
group (`os.setpgrp()` in fork-child) + tree-kill
via `os.killpg(pgid, SIGKILL)` on teardown,
(2) `PR_SET_CHILD_SUBREAPER` on root, (3) explicit
`/proc/<pid>/task/*/children` walk. Vote: (1) —
POSIX-standard, aligns w/ `start_new_session=True`
semantics in `subprocess.Popen` / trio's
`open_process`
- inline reproducer + cleanup recipe scoped to
`$(pwd)/py314/bin/python.*pytest.*spawn-backend=
subint_forkserver` so cleanup doesn't false-flag
unrelated tractor procs (consistent w/
`run-tests` skill's zombie-check guidance)
Stopgap hygiene fix (wiring `reg_addr` through the 5
leaky tests in `test_cancellation.py`) is incoming as
a follow-up — that one stops the blast radius, but
zombies still accumulate per-run until the real
tree-kill fix lands.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Fork-based backends (esp. `subint_forkserver`) can
leak child actor processes on cancelled / SIGINT'd
test runs; the zombies keep the tractor default
registry (`127.0.0.1:1616` / `/tmp/registry@1616.sock`)
bound, so every subsequent session can't bind and
50+ unrelated tests fail with the same
`TooSlowError` / "address in use" signature. Document
the pre-flight + post-cancel check as a mandatory
step 4.
Deats,
- **primary signal**: `ss -tlnp | grep ':1616'` for a
bound TCP registry listener — the authoritative
check since :1616 is unique to our runtime
- `pgrep -af` scoped to `$(pwd)/py[0-9]*/bin/python.*
_actor_child_main|subint-forkserv` for leftover
actor/forkserver procs — scoped deliberately so we
don't false-flag legit long-running tractor-
embedding apps like `piker`
- `ls /tmp/registry@*.sock` for stale UDS sockets
- scoped cleanup recipe (SIGTERM + SIGKILL sweep
using the same `$(pwd)/py*` pattern, UDS `rm -f`,
re-verify) plus a fallback for when a zombie holds
:1616 but doesn't match the pattern: `ss -tlnp` →
kill by PID
- explicit false-positive warning calling out the
`piker` case (`~/repos/piker/py*/bin/python3 -m
tractor._child ...`) so a bare `pgrep` doesn't lead
to nuking unrelated apps
Goal: short-circuit the "spelunking into test code"
rabbit-hole when the real cause is just a leaked PID
from a prior session, without collateral damage to
other tractor-embedding projects on the same box.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Follow-up to 72d1b901 (was prev commit adding `debug_mode` for
`subint_forkserver`): that commit wired the runtime-side
`subint_forkserver` SpawnSpec-recv gate in `Actor._from_parent`, but the
`subint_forkserver_proc` child-target was still passing
`spawn_method='trio'` to `_trio_main` — so `Actor.pformat()` / log lines
would report the subactor as plain `'trio'` instead of the actual
parent-side spawn mechanism. Flip the label to `'subint_forkserver'`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
The `subint_forkserver` backend's child runtime is trio-native (uses
`_trio_main` + receives `SpawnSpec` over IPC just like `trio`/`subint`),
so `tractor.devx.debug._tty_lock` works in those subactors. Wire the
runtime gates that historically hard-coded `_spawn_method == 'trio'` to
recognize this third backend.
Deats,
- new `_DEBUG_COMPATIBLE_BACKENDS` module-const in `tractor._root`
listing the spawn backends whose subactor runtime is trio-native
(`'trio'`, `'subint_forkserver'`). Both the enable-site
(`_runtime_vars['_debug_mode'] = True`) and the cleanup-site reset
key.
off the same tuple — keep them in lockstep when adding backends
- `open_root_actor`'s `RuntimeError` for unsupported backends now
reports the full compatible-set + the rejected method instead of the
stale "only `trio`" msg.
- `runtime._runtime.Actor._from_parent`'s SpawnSpec-recv gate adds
`'subint_forkserver'` to the existing `('trio', 'subint')` tuple
— fork child-side runtime receives the same SpawnSpec IPC handshake as
the others.
- `subint_forkserver_proc` child-target now passes
`spawn_method='subint_forkserver'` (was hard-coded `'trio'`) so
`Actor.pformat()` / log lines reflect the actual parent-side spawn
mechanism rather than masquerading as plain `trio`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Empirical follow-up to the xfail'd orphan-SIGINT test:
the hang is **not** "trio can't install a handler on a
non-main thread" (the original hypothesis from the
`child_sigint` scaffold commit). On py3.14:
- `threading.current_thread() is threading.main_thread()`
IS True post-fork — CPython re-designates the
fork-inheriting thread as "main" correctly
- trio's `KIManager` SIGINT handler IS installed in the
subactor (`signal.getsignal(SIGINT)` confirms)
- the kernel DOES deliver SIGINT to the thread
But `faulthandler` dumps show the subactor wedged in
`trio/_core/_io_epoll.py::get_events` — trio's
wakeup-fd mechanism (which turns SIGINT into an epoll-wake)
isn't firing. So the `except KeyboardInterrupt` at
`tractor/spawn/_entry.py::_trio_main:164` — the runtime's
intentional "KBI-as-OS-cancel" path — never fires.
Deats,
- new `ai/conc-anal/subint_forkserver_orphan_sigint_hang_issue.md`
(+385 LOC): full writeup — TL;DR, symptom reproducer,
the "intentional cancel path" the bug defeats,
diagnostic evidence (`faulthandler` output +
`getsignal` probe), ruled-out hypotheses
(non-main-thread issue, wakeup-fd inheritance,
KBI-as-trio-check-exception), and fix directions
- `test_orphaned_subactor_sigint_cleanup_DRAFT` xfail
`reason` + test docstring rewritten to match the
refined understanding — old wording blamed the
non-main-thread path, new wording points at the
`epoll_wait` wedge + cross-refs the new conc-anal doc
- `_subint_forkserver` module docstring's
`child_sigint='trio'` bullet updated: now notes trio's
handler is already correctly installed, so the flag may
end up a no-op / doc-only mode once the real root cause
is fixed
Closing the gap aligns with existing design intent (make
the already-designed "KBI-as-OS-cancel" behavior actually
fire), not a new feature.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Add configuration surface for future child-side SIGINT
plumbing in `subint_forkserver_proc` without wiring up the
actual trio-native SIGINT bridge — lifting one entry-guard
clause will flip the `'trio'` branch live once the
underlying fork-prelude plumbing is implemented.
Deats,
- new `ChildSigintMode = Literal['ipc', 'trio']` type +
`_DEFAULT_CHILD_SIGINT = 'ipc'` module-level default.
Docstring block enumerates both:
- `'ipc'` (default, currently the only implemented mode):
no child-side SIGINT handler — `trio.run()` is on the
fork-inherited non-main thread where
`signal.set_wakeup_fd()` is main-thread-only, so
cancellation flows exclusively via the parent's
`Portal.cancel_actor()` IPC path. Known gap: orphan
children don't respond to SIGINT
(`test_orphaned_subactor_sigint_cleanup_DRAFT`)
- `'trio'` (scaffolded only): manual SIGINT → trio-cancel
bridge in the fork-child prelude so external Ctrl-C
reaches stuck grandchildren even w/ a dead parent
- `subint_forkserver_proc` pulls `child_sigint` out of
`proc_kwargs` (matches how `trio_proc` threads config to
`open_process`, keeps `start_actor(proc_kwargs=...)` as
the ergonomic entry point); validates membership + raises
`NotImplementedError` for `'trio'` at the backend-entry
guard
- `_child_target` grows a `match child_sigint:` arm that
slots in the future `'trio'` impl without restructuring
— today only the `'ipc'` case is reachable
- module docstring "Still-open work" list grows a bullet
pointing at this config + the xfail'd orphan-SIGINT test
No behavioral change on the default path — `'ipc'` is the
existing flow. Scaffolding only.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Tier-4 test `test_orphaned_subactor_sigint_cleanup_DRAFT`
documents an empirical SIGINT-delivery gap in the
`subint_forkserver` backend: when the parent dies via
`SIGKILL` (no IPC `Portal.cancel_actor()` possible) and
`SIGINT` is sent to the orphan child, the child DOES NOT
unwind — CPython's default `KeyboardInterrupt` is delivered
to `threading.main_thread()`, whose tstate is dead in the
post-fork child bc fork inherited the worker thread, not
main. Trio running on the fork-inherited worker thread
therefore never observes the signal. Marked
`xfail(strict=True)` so the mark flips to XPASS→fail once
the backend grows explicit SIGINT plumbing.
Deats,
- harness runs the failure-mode sequence out-of-process:
1. harness subprocess runs a fresh Python script
that calls `try_set_start_method('subint_forkserver')`
then opens a root actor + one `sleep_forever` subactor
2. parse `PARENT_READY=<pid>` + `CHILD_PID=<pid>` markers
off harness `stdout` to confirm IPC handshake
completed
3. `SIGKILL` the parent, `proc.wait()` to reap the
zombie (otherwise `os.kill(pid, 0)` keeps reporting
it alive)
4. assert the child survived the parent-reap (i.e. was
actually orphaned, not reaped too) before moving on
5. `SIGINT` the orphan child, poll `os.kill(child_pid, 0)`
every 100ms for up to 10s
- supporting helpers: `_read_marker()` with per-proc
bytes-buffer to carry partial lines across calls,
`_process_alive()` liveness probe via `kill(pid, 0)`
- Linux-only via `platform.system() != 'Linux'` skip —
orphan-reparenting semantics don't generalize to
other platforms
- port offset (`reg_addr[1] + 17`) so the harness listener
doesn't race concurrently-running backend tests
- best-effort `finally:` cleanup: `SIGKILL` any still-alive
pids + `proc.kill()` + bounded `proc.wait()` to avoid
leaking orphans across the session
Also, tier-4 header comment documents the cross-backend
generalization path: applicable to any multi-process
backend (`trio`, `mp_spawn`, `mp_forkserver`,
`subint_forkserver`), NOT to plain `subint` (in-process
subints have no orphan OS-child). Move path: lift
harness into `tests/_orphan_harness.py`, parametrize on
session `_spawn_method`, add
`skipif _spawn_method == 'subint'`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Post-fork `_runtime_vars` reset in `subint_forkserver_proc`
was previously done via direct mutation of
`_state._runtime_vars` from an external module + an inline
default dict duplicating the `_state.py`-internal defaults.
Split the access surface into a pure getter + explicit
setter so the reset call site becomes a one-liner
composition.
Deats `tractor/runtime/_state.py`,
- extract initial values into a module-level
`_RUNTIME_VARS_DEFAULTS: dict[str, Any]` constant; the
live `_runtime_vars` is now initialised from
`dict(_RUNTIME_VARS_DEFAULTS)`
- `get_runtime_vars()` grows a `clear_values: bool = False`
kwarg. When True, returns a fresh copy of
`_RUNTIME_VARS_DEFAULTS` instead of the live dict —
still a **pure read**, never mutates anything
- new `set_runtime_vars(rtvars: dict | RuntimeVars)` —
atomic replacement of the live dict's contents via
`.clear()` + `.update()`, so existing references to the
same dict object remain valid. Accepts either the
historical dict form or the `RuntimeVars` struct
Deats `tractor/spawn/_subint_forkserver.py`,
- collapse the prior ad-hoc `.update({...})` block into
`set_runtime_vars(get_runtime_vars(clear_values=True))`
- drop the `_state._current_actor = None` line —
`_trio_main` unconditionally overwrites it downstream,
so no explicit reset needed (noted in the XXX comment)
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
`os.fork()` inherits the parent's entire memory image,
including `tractor.runtime._state` globals that encode
"this process is the root actor" — `_runtime_vars`'s
`_is_root=True`, pre-populated `_root_mailbox` +
`_registry_addrs`, and the parent's `_current_actor`
singleton.
A fresh `exec`-based child starts with those globals at
their module-level defaults (all falsey/empty). The
forkserver child needs to match that shape BEFORE calling
`_actor_child_main()`, otherwise `Actor.__init__()` takes
the `is_root_process() == True` branch and pre-populates
`self.enable_modules`, which then trips
`assert not self.enable_modules` at the top of
`Actor._from_parent()` on the subsequent parent→child
`SpawnSpec` handshake.
Fix: at the start of `_child_target`, null
`_state._current_actor` and overwrite `_runtime_vars` with
a cold-root blank (`_is_root=False`, empty mailbox/addrs,
`_debug_mode=False`) before `_actor_child_main()` runs.
Found-via: `test_subint_forkserver_spawn_basic` hitting
the `enable_modules` assert on child-side runtime boot.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Promote `_subint_forkserver` from primitives-only into a
registered spawn backend: `'subint_forkserver'` is now a
`SpawnMethodKey` literal, dispatched via `_methods` to
the new `subint_forkserver_proc()` target, feature-gated
under the existing `subint`-family py3.14+ case, and
selectable via `--spawn-backend=subint_forkserver`.
Deats,
- new `subint_forkserver_proc()` spawn target in
`_subint_forkserver`:
- mirrors `trio_proc()`'s supervision model — real OS
subprocess so `Portal.cancel_actor()` + `soft_kill()`
on graceful teardown, `os.kill(SIGKILL)` on hard-reap
(no `_interpreters.destroy()` race to fuss over bc the
child lives in its own process)
- only real diff from `trio_proc` is the spawn mechanism:
fork from a main-interp worker thread via
`fork_from_worker_thread()` (off-loaded to trio's
thread pool) instead of `trio.lowlevel.open_process()`
- child-side `_child_target` closure runs
`tractor._child._actor_child_main()` with
`spawn_method='trio'` — the child is a regular trio
actor, "subint_forkserver" names how the parent
spawned, not what the child runs
- new `_ForkedProc` class — thin `trio.Process`-compatible
shim around a raw OS pid: `.poll()` via
`waitpid(WNOHANG)`, async `.wait()` off-loaded to a trio
cache thread, `.kill()` via `SIGKILL`, `.returncode`
cached for repeat calls. `.stdin`/`.stdout`/`.stderr`
are `None` (fork-w/o-exec inherits parent FDs; we don't
marshal them) which matches `soft_kill()`'s `is not None`
guards
Also, new backend-tier test
`test_subint_forkserver_spawn_basic` drives the registered
backend end-to-end via `open_root_actor` + `open_nursery` +
`run_in_actor` w/ a trivial portal-RPC round-trip. Uses a
`forkserver_spawn_method` fixture to flip
`_spawn_method`/`_ctx` for the test's duration + restore on
teardown (so other session-level tests don't observe the
global flip). Test module docstring reworked to describe
the three tiers now covered: (1) primitive-level, (2)
parent-trio-driven primitives, (3) full registered backend.
Status: still-open work (tracked on `tractor#379`) doc'd
inline in the module docstring — no cancel/hard-kill stress
coverage yet, child-side subint-hosted root runtime still
future (gated on `msgspec#563`), thread-hygiene audit
pending the same unblock.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Follow-up tracker companion to the module-docstring TODO
added in `372a0f32`. Catalogs why `_subint_forkserver`'s
two "non-trio thread" constraints
(`fork_from_worker_thread()` +
`run_subint_in_worker_thread()` both allocating dedicated
`threading.Thread`s; test helper named
`run_fork_in_non_trio_thread`) exist today, and which of
them would dissolve once msgspec PEP 684 support ships
(`msgspec#563`) and tractor flips to isolated-mode subints.
Deats,
- three reasons enumerated for the current constraints:
- class-A GIL-starvation — **fixed** by isolated mode:
subints don't share main's GIL so abandoned-thread
contention disappears
- destroy race / tstate-recycling from `subint_proc` —
**unclear**: `_PyXI_Enter` + `_PyXI_Exit` are
cross-mode, so isolated doesn't obviously fix it;
needs empirical retest on py3.14 + isolated API
- fork-from-main-interp-tstate (the CPython-level
`_PyInterpreterState_DeleteExceptMain` gate) — the
narrow reason for using a dedicated thread; **probably
fixed** IF the destroy-race also resolves (bc trio's
cache threads never drove subints → clean main-interp
tstate)
- TL;DR table of which constraints unwind under each
resolution branch
- four-step audit plan for when `msgspec#563` lands:
- flip `_subint` to isolated mode
- empirical destroy-race retest
- audit `_subint_forkserver.py` — drop `non_trio`
qualifier / maybe inline primitives
- doc fallout — close the three `subint_*_issue.md`
siblings w/ post-mortem notes
Also, cross-refs the three sibling `conc-anal/` docs, PEPs
684 + 734, `msgspec#563`, and `tractor#379` (the overall
subint spawn-backend tracking issue).
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
New pytest module `tests/spawn/test_subint_forkserver.py`
drives the forkserver primitives from inside a real
`trio.run()` in the parent — the runtime shape tractor will
actually use when we wire up a `subint_forkserver` spawn
backend proper. Complements the standalone no-trio-in-parent
`ai/conc-anal/subint_fork_from_main_thread_smoketest.py`.
Deats,
- new test pkg `tests/spawn/` (+ empty `__init__.py`)
- two tests, both `@pytest.mark.timeout(30, method='thread')`
for the GIL-hostage safety reason doc'd in
`ai/conc-anal/subint_sigint_starvation_issue.md`:
- `test_fork_from_worker_thread_via_trio` — parent-side
plumbing baseline. `trio.run()` off-loads forkserver
prims via `trio.to_thread.run_sync()` + asserts the
child reaps cleanly
- `test_fork_and_run_trio_in_child` — end-to-end: forked
child calls `run_subint_in_worker_thread()` with a
bootstrap str that does `trio.run()` in a fresh subint
- both tests wrap the inner `trio.run()` in a
`dump_on_hang()` for post-mortem if the outer
`pytest-timeout` fires
- intentionally NOT using `--spawn-backend` — the tests
drive the primitives directly rather than going through
tractor's spawn-method registry (which the forkserver
isn't plugged into yet)
Also, rename `run_trio_in_subint()` →
`run_subint_in_worker_thread()` for naming consistency with
the sibling `fork_from_worker_thread()`. The action is really
"host a subint on a worker thread", not specifically "run
trio" — trio just happens to be the typical payload.
Propagate the rename to the smoketest.
Further, add a "TODO — cleanup gated on msgspec PEP 684
support" section to the `_subint_forkserver` module
docstring: flags the dedicated-`threading.Thread` design as
potentially-revisable once isolated-mode subints are viable
in tractor. Cross-refs `msgspec#563` + `tractor#379` and
points at an audit-plan conc-anal doc we'll add next.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
The smoketest (prior commit) empirically validated the
"fork-from-main-interp-worker-thread" arch on py3.14. Promote
the validated primitives out of the `ai/conc-anal/` smoketest
into `tractor.spawn._subint_forkserver` so they can eventually
be wired into a real "subint forkserver" spawn backend.
Deats,
- new module `tractor/spawn/_subint_forkserver.py` (337 LOC):
- `fork_from_worker_thread(child_target, thread_name)` —
spawn a main-interp `threading.Thread`, call `os.fork()`
from it, shuttle the child pid back to main via a pipe
- `run_trio_in_subint(bootstrap, ...)` — post-fork helper:
create a fresh subint + drive `_interpreters.exec()` on
a dedicated worker thread running the `bootstrap` str
(typically imports `trio`, defines an async entry, calls
`trio.run()`)
- `wait_child(pid, expect_exit_ok)` — `os.waitpid()` +
pass/fail classification reusable from harness AND the
eventual real spawn path
- feature-gated py3.14+ via the public
`concurrent.interpreters` presence check; matches the gate
in `tractor.spawn._subint`
- module docstring doc's the CPython-block context
(cross-refs `_subint_fork` stub + the two `conc-anal/`
docs) and status: EXPERIMENTAL, not yet registered in
`_spawn._methods`
Also, refactor the smoketest
`ai/conc-anal/subint_fork_from_main_thread_smoketest.py` to
import the primitives from the new module rather than inline
its own copies. Keeps the smoketest and the tractor-side
impl in sync as the forkserver design evolves; the smoketest
remains a zero-`tractor`-runtime CPython-level check
(imports ONLY the three primitives, no runtime bring-up).
Status: next step is to drive these from a parent-side
`trio.run()` and hook the returned child pid into the normal
actor-nursery/IPC flow — then register `subint_forkserver`
as a `SpawnMethodKey` in `_spawn.py`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Standalone script to validate the "main-interp worker-thread
forkserver + subint-hosted trio" arch proposed as a workaround
to the CPython-level refusal doc'd in
`ai/conc-anal/subint_fork_blocked_by_cpython_post_fork_issue.md`.
Deliberately NOT a `tractor` test — zero `tractor` imports.
Uses `_interpreters` (private stdlib) + `os.fork()` directly so
pass/fail is a property of CPython alone, independent of our
runtime. Requires py3.14+.
Deats,
- four scenarios via `--scenario`:
- `control_subint_thread_fork` — the KNOWN-BROKEN case as a
harness sanity; if the child DOESN'T abort, our analysis
is wrong
- `main_thread_fork` — baseline sanity, must always succeed
- `worker_thread_fork` — architectural assertion: regular
`threading.Thread` attached to main interp calls
`os.fork()`; child should survive post-fork cleanup
- `full_architecture` — end-to-end: fork from a main-interp
worker thread, then in child create a subint driving a
worker thread running `trio.run()`
- exit code 0 on EXPECTED outcome (for `control_*` that means
"child aborted", not "child succeeded")
- each scenario prints a self-contained pass/fail banner; use
`os.waitpid()` of the parent + per-scenario status prints to
observe the child's fate
Also, log NLNet provenance for this session's three-sub-phase
work (py3.13 gate tightening, `pytest-timeout` + marker
refactor, `subint_fork` prototype → CPython-block finding).
Prompt-IO: ai/prompt-io/claude/20260422T200723Z_797f57c_prompt_io.md
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Empirical finding: the WIP `subint_fork_proc` scaffold
landed in `cf0e3e6f` does *not* work on current CPython.
The `fork()` syscall succeeds in the parent, but the
CHILD aborts immediately during
`PyOS_AfterFork_Child()` →
`_PyInterpreterState_DeleteExceptMain()`, which gates
on the current tstate belonging to the main interp —
the child dies with `Fatal Python error: not main
interpreter`.
CPython devs acknowledge the fragility with an in-source
comment (`// Ideally we could guarantee tstate is running
main.`) but expose no user-facing hook to satisfy the
precondition — so the strategy is structurally dead until
upstream changes.
Rather than delete the scaffold, reshape it into a
documented dead-end so the next person with this idea
lands on the reason rather than rediscovering the same
CPython-level refusal.
Deats,
- Move `subint_fork_proc` out of `tractor.spawn._subint`
into a new `tractor.spawn._subint_fork` dedicated
module (153 LOC). Module + fn docstrings now describe
the blockage directly; the fn body is trimmed to a
`NotImplementedError` pointing at the analysis doc —
no more dead-code `bootstrap` sketch bloating
`_subint.py`.
- `_spawn.py`: keep `'subint_fork'` in `SpawnMethodKey`
+ the `_methods` dispatch so
`--spawn-backend=subint_fork` routes to a clean
`NotImplementedError` rather than "invalid backend";
comment calls out the blockage. Collapse the duplicate
py3.14 feature-gate in `try_set_start_method()` into a
combined `case 'subint' | 'subint_fork':` arm.
- New 337-line analysis:
`ai/conc-anal/subint_fork_blocked_by_cpython_post_fork_issue.md`.
Annotated walkthrough from the user-visible fatal
error down to the specific `Modules/posixmodule.c` +
`Python/pystate.c` source lines enforcing the refusal,
plus an upstream-report draft.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Experimental third spawn backend: use a fresh
sub-interpreter purely as a trio-free launchpad from
which to `os.fork()` + exec back into
`python -m tractor._child`. Per issue #379's
"fork()-workaround/hacks" thread.
Intent is to sidestep both,
- the trio+fork hazards hitting `trio_proc` (python- trio/trio#1614 et
al.), since the forking interp is guaranteed trio-free.
- the shared-GIL abandoned-thread hazards hitting `subint_proc`
(`ai/conc-anal/subint_sigint_starvation_issue.md`), since we don't
*stay* in the subint — it only lives long enough to call `os.fork()`
Downstream of the fork+exec, all the existing `trio_proc` plumbing is
reused verbatim: `ipc_server.wait_for_peer()`, `SpawnSpec`, `Portal`
yield, soft-kill.
Status: NOT wired up beyond scaffolding. The fn raises
`NotImplementedError` immediately; the `bootstrap` fork/exec string
builder and the `# TODO: orchestrate driver thread` block are kept
in-tree as deliberate dead code so the next iteration starts from
a concrete shape rather than a blank page.
Docstring calls out three open questions that need
empirical validation before wiring this up:
1. Does CPython permit `os.fork()` from a non-main
legacy subint?
2. Can the child stay fork-without-exec and
`trio.run()` directly from within the launchpad
subint?
3. How do `signal.set_wakeup_fd()` handlers and other
process-global state interact when the forking
thread is inside a subint?
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Adopt the `@pytest.mark.skipon_spawn_backend('subint',
reason=...)` marker (a617b521) across the suites
reproducing the `subint` GIL-contention / starvation
hang classes doc'd in `ai/conc-anal/subint_*_issue.md`.
Deats,
- Module-level `pytestmark` on full-file-hanging suites:
- `tests/test_cancellation.py`
- `tests/test_inter_peer_cancellation.py`
- `tests/test_pubsub.py`
- `tests/test_shm.py`
- Per-test decorator where only one test in the file
hangs:
- `tests/discovery/test_registrar.py
::test_stale_entry_is_deleted` — replaces the
inline `if start_method == 'subint': pytest.skip`
branch with a declarative skip.
- `tests/test_subint_cancellation.py
::test_subint_non_checkpointing_child`.
- A few per-test decorators are left commented-in-
place as breadcrumbs for later finer-grained unskips.
Also, some nearby tidying in the affected files:
- Annotate loose fixture / test params
(`pytest.FixtureRequest`, `str`, `tuple`, `bool`) in
`tests/conftest.py`, `tests/devx/conftest.py`, and
`tests/test_cancellation.py`.
- Normalize `"""..."""` → `'''...'''` docstrings per
repo convention on a few touched tests.
- Add `timeout=6` / `timeout=10` to
`@tractor_test(...)` on `test_cancel_infinite_streamer`
and `test_some_cancels_all`.
- Drop redundant `spawn_backend` param from
`test_cancel_via_SIGINT`; use `start_method` in the
`'mp' in ...` check instead.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
A reusable `@pytest.mark.skipon_spawn_backend( '<backend>' [, ...],
reason='...')` marker for backend-specific known-hang / -borked cases
— avoids scattering `@pytest.mark.skipif(lambda ...)` branches across
tests that misbehave under a particular `--spawn-backend`.
Deats,
- `pytest_configure()` registers the marker via
`addinivalue_line('markers', ...)`.
- New `pytest_collection_modifyitems()` hook walks
each collected item with `item.iter_markers(
name='skipon_spawn_backend')`, checks whether the
active `--spawn-backend` appears in `mark.args`, and
if so injects a concrete `pytest.mark.skip(
reason=...)`. `iter_markers()` makes the decorator
work at function, class, or module (`pytestmark =
[...]`) scope transparently.
- First matching mark wins; default reason is
`f'Borked on --spawn-backend={backend!r}'` if the
caller doesn't supply one.
Also, tighten type annotations on nearby `pytest`
integration points — `pytest_configure`, `debug_mode`,
`spawn_backend`, `tpt_protos`, `tpt_proto` — now taking
typed `pytest.Config` / `pytest.FixtureRequest` params.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Add two more tests to the catalog in
`conc-anal/subint_sigint_starvation_issue.md` — same
signal-wakeup-fd-saturation fingerprint (abandoned legacy-subint driver
threads → shared-GIL starvation → `write() = EAGAIN` on the wakeup pipe
→ silent SIGINT drop), different load patterns.
Deats,
- `test_cancel_while_childs_child_in_sync_sleep[subint-False]`: nested
actor-tree + sync-sleeping grandchild. Under `trio`/`mp_*` the "zombie
reaper" is a subproc `SIGKILL`; no equivalent exists under subint, so
the grandchild persists in its abandoned driver thread. Often only
manifests under full-suite runs (earlier tests seed the
abandoned-thread pool).
- `test_multierror_fast_nursery[subint-25-0.5]`: 25 concurrent subactors
all go through teardown on the multierror. Bounded hard-kills run in
parallel — so the total budget is ~3s, not 3s × 25. Leaves 25
abandoned driver threads simultaneously alive, an extreme pressure
multiplier. `strace` shows several successful `write(16, "\2", 1) = 1`
(GIL round-robin IS giving main brief slices) before finally
saturating with `EAGAIN`.
Also include a `pstree -snapt <pid>` capture showing
16+ live `{subint-driver[<interp_id>}` threads at the
moment of hang — the direct GIL-contender population.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Add a hard process-level wall-clock bound on the two
known-hanging subint-backend tests so an unattended
suite run can't wedge indefinitely in either of the
hang classes doc'd in `ai/conc-anal/`.
Deats,
- New `testing` dep: `pytest-timeout>=2.3`.
- `test_stale_entry_is_deleted`:
`@pytest.mark.timeout(3, method='thread')`. The
`method='thread'` choice is deliberate —
`method='signal'` routes via `SIGALRM` which is
starved by the same GIL-hostage path that drops
`SIGINT` (see `subint_sigint_starvation_issue.md`),
so it'd never actually fire in the starvation case.
- `test_subint_non_checkpointing_child`: same
decorator, same reasoning (defense-in-depth over
the inner `trio.fail_after(15)`).
At timeout, `pytest-timeout` hard-kills the pytest
process itself — that's the intended behavior here;
the alternative is the suite never returning.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Log the `claude-opus-4-7` collab that produced `e92e3cd2` ("Doc `subint`
backend hang classes + arm `dump_on_hang`"). Substantive bc the two new
`ai/conc-anal/` docs were jointly authored — user framed the two-class
split + set candidate-fix ordering for the class-2 (Ctrl-C-able) hang;
claude drafted the prose and the test-side cross-linking comments.
`.raw.md` is in diff-ref mode — per-file pointers via `git diff
e92e3cd2~1..e92e3cd2 -- <path>` rather than re-embedding content that
already lives in `git log -p`.
Prompt-IO: ai/prompt-io/claude/20260420T192739Z_5e8cd8b2_prompt_io.md
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Classify and write up the two distinct hang modes hit during Phase
B subint bringup (issue #379) so future triage doesn't re-derive them
from scratch.
Deats, two new `ai/conc-anal/` docs,
- `subint_sigint_starvation_issue.md`: abandoned legacy-subint thread
+ shared GIL → main trio loop starves → signal-wakeup-fd pipe fills
→ `SIGINT` silently dropped (`strace` shows `write() = EAGAIN` on the
wakeup-fd). Un- Ctrl-C-able. Structurally a CPython limit; blocked on
`msgspec` PEP 684 (jcrist/msgspec#563)
- `subint_cancel_delivery_hang_issue.md`: parent-side trio task parks on
an orphaned IPC channel after subint teardown — no clean EOF delivered
to the waiting receive. Ctrl-C-able (main loop iterates fine); OUR bug
to fix. Candidate fix: explicit parent-side channel abort in
`subint_proc`'s hard-kill teardown
Cross-link the docs from their test reproducers,
- `test_stale_entry_is_deleted` (→ starvation class): wrap
`trio.run(main)` in `dump_on_hang(seconds=20)` so a future regression
captures a stack dump. Kept un- skipped so the dump file is
inspectable
- `test_subint_non_checkpointing_child` (→ delivery class): extend
docstring with a "KNOWN ISSUE" block pointing at the analysis
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Lock in the escape-hatch machinery added to `tractor.spawn._subint`
during the Phase B.2/B.3 bringup (issue #379) so future stdlib
regressions or our own refactors don't silently re-introduce the
mid-suite hangs.
Deats,
- `test_subint_happy_teardown`: baseline — spawn a subactor, one portal
RPC, clean teardown. If this breaks, something's wrong unrelated to
the hard-kill shields.
- `test_subint_non_checkpointing_child`: cancel a subactor stuck in
a non-checkpointing Python loop (`threading.Event.wait()` releases the
GIL but never inserts a trio checkpoint). Validates the bounded-shield
+ daemon-driver-thread combo abandons the thread after
`_HARD_KILL_TIMEOUT`.
Every test is wrapped in `trio.fail_after()` for a deterministic
per-test wall-clock ceiling (an unbounded audit would defeat itself) and
arms `tractor.devx.dump_on_hang()` so a hang captures a stack dump
— pytest's stderr capture swallows `faulthandler` output by default.
Gated via `pytest.importorskip('concurrent.interpreters')` and
a module-level skip when `--spawn-backend` isn't `'subint'`.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
The private `_interpreters` C module ships since 3.13, but that vintage
wedges under our `threading.Thread` + multi-trio usage pattern
—> `_interpreters.exec()` silently never makes progress. 3.14 fixes it.
So gate on the presence of the public `concurrent.interpreters` wrapper
(3.14+ only) even tho we still call into the private module at runtime.
Deats,
- `try_set_start_method('subint')` error msg + `_subint` module
docstring/comments rewritten to document the 3.14 floor and why 3.13
can't work.
- `_subint._has_subints` gate now imports `concurrent.interpreters` (not
`_interpreters`) as the version sentinel.
Also, reshuffle `pyproject.toml` deps into
per-python-version `[tool.uv.dependency-groups]`:
- `subints` group: `msgspec>=0.21.0`, py>=3.14
- `eventfd` group: `cffi>=1.17.1`, py>=3.13,<3.14
- `sync_pause` group: `greenback`, py>=3.13,<3.14
(was in `devx`; moved out bc no 3.14 yet)
Bump top-level `msgspec>=0.20.0` too.
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Bottle up the diagnostic primitives that actually cracked the
silent mid-suite hangs in the `subint` spawn-backend bringup (issue
there" session has them on the shelf instead of reinventing from
scratch.
Deats,
- `dump_on_hang(seconds, *, path)` — context manager wrapping
`faulthandler.dump_traceback_later()`. Critical gotcha baked in:
dumps go to a *file*, not `sys.stderr`, bc pytest's stderr
capture silently eats the output and you can spend an hour
convinced you're looking at the wrong thing
- `track_resource_deltas(label, *, writer)` — context manager
logging per-block `(threading.active_count(),
len(_interpreters.list_all()))` deltas; quickly rules out
leak-accumulation theories when a suite progressively worsens (if
counts don't grow, it's not a leak, look for a race on shared
cleanup instead)
- `resource_delta_fixture(*, autouse, writer)` — factory returning
a `pytest` fixture wrapping `track_resource_deltas` per-test; opt
in by importing into a `conftest.py`. Kept as a factory (not a
bare fixture) so callers own `autouse` / `writer` wiring
Also,
- export the three names from `tractor.devx`
- dep-free on py<3.13 (swallows `ImportError` for `_interpreters`)
- link back to the provenance in the module docstring (issue #379 /
commit `26fb820`)
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Unbounded `trio.CancelScope(shield=True)` at the
soft-kill and thread-join sites can wedge the parent
trio loop indefinitely when a stuck subint ignores
portal-cancel (e.g. bc the IPC channel is already
broken).
Deats,
- add `_HARD_KILL_TIMEOUT` (3s) module-level const
- wrap both shield sites with
`trio.move_on_after()` so we abandon a stuck
subint after the deadline
- flip driver thread to `daemon=True` so proc-exit
also isn't blocked by a wedged subint
- pass `abandon_on_cancel=True` to
`trio.to_thread.run_sync(driver_thread.join)`
— load-bearing for `move_on_after` to actually
fire
- log warnings when either timeout triggers
- improve `InterpreterError` log msg to explain
the abandoned-thread scenario
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Log the `claude-opus-4-7` session that produced
the `_subint.py` dedicated-thread fix (`26fb8206`).
Substantive bc the patch was entirely AI-generated;
raw log also preserves the CPython-internals
research informing Phase B.3 hard-kill work.
Prompt-IO: ai/prompt-io/claude/20260418T042526Z_26fb820_prompt_io.md
(this commit msg was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
`trio.to_thread.run_sync(_interpreters.exec, ...)` runs `exec()` on
a cached worker thread — and when that thread is returned to the
cache after the subint's `trio.run()` exits, CPython still keeps
the subint's tstate attached to the (now idle) worker. Result: the
teardown `_interpreters.destroy(interp_id)` in the `finally` block
can block the parent's trio loop indefinitely, waiting for a tstate
release that only happens when the worker either picks up a new job
or exits.
Manifested as intermittent mid-suite hangs under
`--spawn-backend=subint` — caught by a
`faulthandler.dump_traceback_later()` showing the main thread stuck
in `_interpreters.destroy()` at `_subint.py:293` with only an idle
trio-cache worker as the other live thread.
Deats,
- drive the subint on a plain `threading.Thread` (not
`trio.to_thread`) so the OS thread truly exits after
`_interpreters.exec()` returns, releasing tstate and unblocking
destroy
- signal `subint_exited.set()` back to the parent trio loop from
the driver thread via `trio.from_thread.run_sync(...,
trio_token=...)` — capture the token at `subint_proc` entry
- swallow `trio.RunFinishedError` in that signal path for the case
where parent trio has already exited (proc teardown)
- in the teardown `finally`, off-load the sync
`driver_thread.join()` to `trio.to_thread.run_sync` (cache thread
w/ no subint tstate → safe) so we actually wait for the driver to
exit before `_interpreters.destroy()`
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Expand the comment block above the `_interpreters`
import explaining *why* we use the private C mod
over `concurrent.interpreters`: the public API only
exposes PEP 734's `'isolated'` config which breaks
`msgspec` (missing PEP 684 slot). Add reference
links to PEP 734, PEP 684, cpython sources, and
the msgspec upstream tracker (jcrist/msgspec#563).
Also,
- update error msgs in both `_spawn.py` and
`_subint.py` to say "3.13+" (matching the actual
`_interpreters` availability) instead of "3.14+".
- tweak the mod docstring to reflect py3.13+
availability via the private C module.
Review: PR #444 (copilot-pull-request-reviewer)
https://github.com/goodboy/tractor/pull/444
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Gud Boi