Files
codex/codex-rs/utils/pty/src/process_group.rs
T
viyatb-oai 9152ebd289 fix(linux-sandbox): preserve shell cleanup on interruption (#22729)
## Why
Interrupted `shell_command` calls can race with the outer tool-dispatch
cancellation path. When that happens, the runtime future may be dropped
before the spawned process gets a chance to run `SIGTERM` cleanup. For
bwrapd-backed Linux sandbox commands, that can leave synthetic
protected-path mount bookkeeping such as `.git/.codex` registrations
under `/tmp` behind after a TUI interruption.

The relevant cancellation points are the outer dispatch race in
[`core/src/tools/parallel.rs`](https://github.com/openai/codex/blob/bd184ba84703cc924921ed883f0cf17d3dba60ff/codex-rs/core/src/tools/parallel.rs#L91-L132)
and the process shutdown logic in
[`core/src/exec.rs`](https://github.com/openai/codex/blob/bd184ba84703cc924921ed883f0cf17d3dba60ff/codex-rs/core/src/exec.rs#L1367-L1393).

## What changed
- Keep `shell_command` dispatch alive long enough for the runtime to
finish cancellation cleanup instead of immediately returning the
synthetic aborted response.
- Fold shell-turn cancellation into the existing `ExecExpiration` path
in
[`core/src/tools/runtimes/shell.rs`](https://github.com/openai/codex/blob/bd184ba84703cc924921ed883f0cf17d3dba60ff/codex-rs/core/src/tools/runtimes/shell.rs#L267-L274),
so cancellation and timeout behavior stay centralized.
- On cancellation, send `SIGTERM` first, wait briefly for cleanup to
run, then hard-kill any remaining descendants in the original process
group.
- Treat `ESRCH` as an already-gone process-group cleanup case in
`codex-utils-pty`, which keeps best-effort teardown from surfacing a
stale-process race as an error.

## Verification
- `cargo test -p codex-core cancellation`
- Added regression coverage for:
  - `shell_tool_cancellation_waits_for_runtime_cleanup`
  - `process_exec_tool_call_cancellation_allows_sigterm_cleanup`
2026-05-27 12:59:11 -07:00

185 lines
5.2 KiB
Rust

//! Process-group helpers shared by pipe/pty and shell command execution.
//!
//! This module centralizes the OS-specific pieces that ensure a spawned
//! command can be cleaned up reliably:
//! - `set_process_group` is called in `pre_exec` so the child starts its own
//! process group.
//! - `detach_from_tty` starts a new session so non-interactive children do not
//! inherit the controlling TTY.
//! - `kill_process_group_by_pid` targets the whole group (children/grandchildren)
//! - `kill_process_group` targets a known process group ID directly
//! instead of a single PID.
//! - `set_parent_death_signal` (Linux only) arranges for the child to receive a
//! `SIGTERM` when the parent exits, and re-checks the parent PID to avoid
//! races during fork/exec.
//!
//! On non-Unix platforms these helpers are no-ops.
use std::io;
use tokio::process::Child;
#[cfg(target_os = "linux")]
/// Ensure the child receives SIGTERM when the original parent dies.
///
/// This should run in `pre_exec` and uses `parent_pid` captured before spawn to
/// avoid a race where the parent exits between fork and exec.
pub fn set_parent_death_signal(parent_pid: libc::pid_t) -> io::Result<()> {
if unsafe { libc::prctl(libc::PR_SET_PDEATHSIG, libc::SIGTERM) } == -1 {
return Err(io::Error::last_os_error());
}
if unsafe { libc::getppid() } != parent_pid {
unsafe {
libc::raise(libc::SIGTERM);
}
}
Ok(())
}
#[cfg(not(target_os = "linux"))]
/// No-op on non-Linux platforms.
pub fn set_parent_death_signal(_parent_pid: i32) -> io::Result<()> {
Ok(())
}
#[cfg(unix)]
/// Detach from the controlling TTY by starting a new session.
pub fn detach_from_tty() -> io::Result<()> {
let result = unsafe { libc::setsid() };
if result == -1 {
let err = io::Error::last_os_error();
if err.raw_os_error() == Some(libc::EPERM) {
return set_process_group();
}
return Err(err);
}
Ok(())
}
#[cfg(not(unix))]
/// No-op on non-Unix platforms.
pub fn detach_from_tty() -> io::Result<()> {
Ok(())
}
#[cfg(unix)]
/// Put the calling process into its own process group.
///
/// Intended for use in `pre_exec` so the child becomes the group leader.
pub fn set_process_group() -> io::Result<()> {
let result = unsafe { libc::setpgid(0, 0) };
if result == -1 {
Err(io::Error::last_os_error())
} else {
Ok(())
}
}
#[cfg(not(unix))]
/// No-op on non-Unix platforms.
pub fn set_process_group() -> io::Result<()> {
Ok(())
}
#[cfg(unix)]
/// Kill the process group for the given PID (best-effort).
///
/// This resolves the PGID for `pid` and sends SIGKILL to the whole group.
pub fn kill_process_group_by_pid(pid: u32) -> io::Result<()> {
use std::io::ErrorKind;
let pid = pid as libc::pid_t;
let pgid = unsafe { libc::getpgid(pid) };
if pgid == -1 {
let err = io::Error::last_os_error();
if err.kind() != ErrorKind::NotFound && err.raw_os_error() != Some(libc::ESRCH) {
return Err(err);
}
return Ok(());
}
let result = unsafe { libc::killpg(pgid, libc::SIGKILL) };
if result == -1 {
let err = io::Error::last_os_error();
if err.kind() != ErrorKind::NotFound && err.raw_os_error() != Some(libc::ESRCH) {
return Err(err);
}
}
Ok(())
}
#[cfg(not(unix))]
/// No-op on non-Unix platforms.
pub fn kill_process_group_by_pid(_pid: u32) -> io::Result<()> {
Ok(())
}
#[cfg(unix)]
/// Send SIGTERM to a specific process group ID (best-effort).
///
/// Returns `Ok(true)` when SIGTERM was delivered to an existing group and
/// `Ok(false)` when the group no longer exists.
pub fn terminate_process_group(process_group_id: u32) -> io::Result<bool> {
use std::io::ErrorKind;
let pgid = process_group_id as libc::pid_t;
let result = unsafe { libc::killpg(pgid, libc::SIGTERM) };
if result == -1 {
let err = io::Error::last_os_error();
if err.kind() == ErrorKind::NotFound || err.raw_os_error() == Some(libc::ESRCH) {
return Ok(false);
}
return Err(err);
}
Ok(true)
}
#[cfg(not(unix))]
/// No-op on non-Unix platforms.
pub fn terminate_process_group(_process_group_id: u32) -> io::Result<bool> {
Ok(false)
}
#[cfg(unix)]
/// Kill a specific process group ID (best-effort).
pub fn kill_process_group(process_group_id: u32) -> io::Result<()> {
use std::io::ErrorKind;
let pgid = process_group_id as libc::pid_t;
let result = unsafe { libc::killpg(pgid, libc::SIGKILL) };
if result == -1 {
let err = io::Error::last_os_error();
if err.kind() != ErrorKind::NotFound && err.raw_os_error() != Some(libc::ESRCH) {
return Err(err);
}
}
Ok(())
}
#[cfg(not(unix))]
/// No-op on non-Unix platforms.
pub fn kill_process_group(_process_group_id: u32) -> io::Result<()> {
Ok(())
}
#[cfg(unix)]
/// Kill the process group for a tokio child (best-effort).
pub fn kill_child_process_group(child: &mut Child) -> io::Result<()> {
if let Some(pid) = child.id() {
return kill_process_group_by_pid(pid);
}
Ok(())
}
#[cfg(not(unix))]
/// No-op on non-Unix platforms.
pub fn kill_child_process_group(_child: &mut Child) -> io::Result<()> {
Ok(())
}