Cross-process named locks for Rust.
Works across both threads and processes on the same machine. The
cross-platform code uses no unsafe; platform-specific Win32 calls are
contained in a small, scoped module.
MSRV: Rust 1.89+
| Platform | Mechanism |
|---|---|
| Unix (Linux, macOS, …) | flock(2) via std::fs::File::lock() |
| Windows | CreateMutexW (kernel named mutex, Global\ namespace) |
[dependencies]
ipc-lock = "0.1"use ipc_lock::{Lock, Result};
fn main() -> Result<()> {
let lock = Lock::new("my-app")?;
let _guard = lock.lock()?; // blocks until the lock is free
// critical section …
Ok(()) // _guard dropped → lock released automatically
}Lock is cheaply cloneable — all clones share the same underlying state:
let lock = Lock::new("my-app")?;
let other = lock.clone(); // O(1) Arc clone, same lock
let _guard = lock.lock()?;
assert!(other.try_lock().is_err()); // WouldBlock- Must not be empty.
- Must not contain
\0,/, or\.
On Unix the lock file is placed at $TMPDIR/<name>.lock (falling back to
/tmp/<name>.lock). Use [Lock::with_path] to specify an exact path.
The library intentionally leaves the lock file in place after the lock is
released. You can retrieve the path with [Lock::path] and delete it when you
know no other process is using the lock:
let lock = Lock::new("my-app")?;
// ... use the lock ...
let path = lock.path().to_owned();
drop(lock);
std::fs::remove_file(path)?;Deleting the file while another process may still be using the lock can break mutual exclusion, because a new process would create a fresh file at the same path.
If a Windows process terminates without releasing the named mutex, the next
waiter still acquires the lock successfully, but
[LockGuard::is_abandoned()] returns true. This signals that any shared
state protected by the lock may be inconsistent.
MIT — see LICENSE.