gh-149527: fix creation of temporary socket files post GH-148578 (#149597)

Author: picnixz

Lib/multiprocessing/connection.py

@@ -76,11 +76,12 @@ def arbitrary_address(family):
     if family == 'AF_INET':
         return ('localhost', 0)
     elif family == 'AF_UNIX':
-        # NOTE: util.get_temp_dir() is a 0o700 per-process directory. A
-        # mktemp-style ToC vs ToU concern is not important; bind() surfaces
-        # the extremely unlikely collision as EADDRINUSE.
+        # NOTE: util.get_temp_dir() is a 0o700 per-process directory.
+        # A mktemp-style ToC vs ToU concern is not important as bind()
+        # surfaces the extremely unlikely collision as EADDRINUSE.
+        suffix = os.urandom(util._TMPSOCK_SUFFIXLEN // 2).hex()
         return os.path.join(util.get_temp_dir(),
-                            f'sock-{os.urandom(6).hex()}')
+                            f"{util._TMPSOCK_PREFIX}{suffix}")
     elif family == 'AF_PIPE':
         return (r'\\.\pipe\pyc-%d-%d-%s' %
                 (os.getpid(), next(_mmap_counter), os.urandom(8).hex()))

Lib/multiprocessing/util.py

@@ -143,6 +143,41 @@ def is_abstract_socket_namespace(address):
     # On Windows platforms, we do not create AF_UNIX sockets.
     _SUN_PATH_MAX = None if os.name == 'nt' else 92
 
+# The temporary socket file path (without NULL terminator) is
+#
+#   TEMPDIR/{_TMPPYMP_PREFIX}{S1}/{_TMPSOCK_PREFIX}{S2}
+#
+# where
+#
+# - S1 is a random suffix of length 8 generated by get_temp_dir().
+#   This is used to generate the temporary sub-directory that will
+#   contains the socket file. Update _TMPPYMP_SUFFIXLEN accordingly
+#   if the random suffix length chosen by tempfile.mkdtemp() changes.
+#
+# - S2 is a random suffix of length 2N generated by os.urandom(N).hex()
+#   in multiprocessing.connection.arbitrary_address(). The value of N
+#   is not fixed and may vary across Python versions as to minimize
+#   collisions and path lengths.
+#
+_TMPPYMP_PREFIX = "pymp-"
+_TMPPYMP_PREFIXLEN = len(_TMPPYMP_PREFIX)
+_TMPPYMP_SUFFIXLEN = 8
+_TMPSOCK_PREFIX = "sock-"
+_TMPSOCK_PREFIXLEN = len(_TMPSOCK_PREFIX)
+_TMPSOCK_SUFFIXLEN = 12
+
+# Length of the socket filepath from the chosen temporary directory,
+# including a leading path separator and the path separator between
+# the temporary subdirectory and the socket filename.
+_SUN_PATH_LEN_RESERVED = (
+    len(os.path.sep) + _TMPPYMP_PREFIXLEN + _TMPPYMP_SUFFIXLEN +
+    len(os.path.sep) + _TMPSOCK_PREFIXLEN + _TMPSOCK_SUFFIXLEN
+)
+assert (
+    _SUN_PATH_MAX is None
+    or _SUN_PATH_LEN_RESERVED < _SUN_PATH_MAX
+)
+
 
 def _remove_temp_dir(rmtree, tempdir):
     rmtree(tempdir)
@@ -161,24 +196,16 @@ def _get_base_temp_dir(tempfile):
     """
     if os.name == 'nt':
         return None
-    # Most of the time, the default temporary directory is /tmp. Thus,
-    # listener sockets files "$TMPDIR/pymp-XXXXXXXX/sock-XXXXXXXX" do
-    # not have a path length exceeding SUN_PATH_MAX.
+    # Most of the time, the default temporary directory is /tmp. As such,
+    # the path length of listener sockets files does not exceed SUN_PATH_MAX.
     #
     # If users specify their own temporary directory, we may be unable
     # to create those files. Therefore, we fall back to the system-wide
     # temporary directory /tmp, assumed to exist on POSIX systems.
     #
     # See https://github.com/python/cpython/issues/132124.
     base_tempdir = tempfile.gettempdir()
-    # Files created in a temporary directory are suffixed by a string
-    # generated by tempfile._RandomNameSequence, which, by design,
-    # is 8 characters long.
-    #
-    # Thus, the socket file path length (without NULL terminator) will be:
-    #
-    #   len(base_tempdir + '/pymp-XXXXXXXX' + '/sock-XXXXXXXX')
-    sun_path_len = len(base_tempdir) + 14 + 14
+    sun_path_len = len(base_tempdir) + _SUN_PATH_LEN_RESERVED
     # Strict inequality to account for the NULL terminator.
     # See https://github.com/python/cpython/issues/140734.
     if sun_path_len < _SUN_PATH_MAX:
@@ -206,8 +233,8 @@ def _get_base_temp_dir(tempfile):
         # not be able to write socket files out there.
         return base_tempdir
     warn("Ignoring user-defined temporary directory: %s", base_tempdir)
-    # at most max(map(len, dirlist)) + 14 + 14 = 36 characters
-    assert len(base_system_tempdir) + 14 + 14 < _SUN_PATH_MAX
+    # The following assertion must be satisfied with the chosen constants.
+    assert len(base_system_tempdir) + _SUN_PATH_LEN_RESERVED < _SUN_PATH_MAX
     return base_system_tempdir
 
 
@@ -218,7 +245,7 @@ def get_temp_dir():
         # Deferred until first use to keep importing this module cheap.
         import shutil, tempfile
         base_tempdir = _get_base_temp_dir(tempfile)
-        tempdir = tempfile.mkdtemp(prefix='pymp-', dir=base_tempdir)
+        tempdir = tempfile.mkdtemp(prefix=_TMPPYMP_PREFIX, dir=base_tempdir)
         info('created temp directory %s', tempdir)
         # keep a strong reference to shutil.rmtree(), since the finalizer
         # can be called late during Python shutdown

Lib/test/test_multiprocessing_forkserver/test_misc.py

@@ -1,7 +1,47 @@
+import os
+import re
 import unittest
+from multiprocessing.util import _SUN_PATH_MAX
 from test._test_multiprocessing import install_tests_in_module_dict
+from test.support import os_helper, script_helper
 
 install_tests_in_module_dict(globals(), 'forkserver', exclude_types=True)
 
+
+class TestForkServerConfiguration(unittest.TestCase):
+    def test_respect_sun_path_max(self):
+        # Ensure that the calculation for temporary filepath lengths is correct.
+        # See https://github.com/python/cpython/issues/149527.
+
+        cmd = '''if 1:
+            from multiprocessing.connection import arbitrary_address
+            from multiprocessing.util import get_temp_dir
+            if __name__ == "__main__":
+                print(get_temp_dir())
+                print(arbitrary_address("AF_UNIX"))
+        '''
+        with os_helper.temp_dir() as root:
+            self.assertLess(len(root), _SUN_PATH_MAX)
+            _, out, _ = script_helper.assert_python_ok('-c', cmd, TMPDIR=root)
+        res = out.decode().strip().splitlines()
+        self.assertEqual(len(res), 2)
+
+        # get_temp_dir() with a TMPDIR that has a too large length ends up
+        # NOT being used as it is not possible to use it. In particular,
+        # the corresponding path will fallback to one of the "built-in"
+        # ones such as '/tmp', '/var/tmp', or '/usr/tmp'.
+        root_regex = "|".join(map(re.escape, [root, "/tmp", "/var/tmp", "/usr/tmp"]))
+
+        temp_pymp = res[0]
+        self.assertLess(len(temp_pymp), _SUN_PATH_MAX)
+        temp_pymp_regex = os.path.join(root_regex, r"pymp-\w{8}")
+        self.assertRegex(temp_pymp, temp_pymp_regex)
+
+        temp_sock = res[1]
+        self.assertLess(len(temp_sock), _SUN_PATH_MAX)
+        temp_sock_regex = os.path.join(temp_pymp_regex, r"sock-[0-9a-fA-F]{12}")
+        self.assertRegex(temp_sock, temp_sock_regex)
+
+
 if __name__ == '__main__':
     unittest.main()

Misc/NEWS.d/next/Library/2026-05-09-12-58-04.gh-issue-149527.hX36Yh.rst

@@ -0,0 +1,3 @@
+:mod:`multiprocessing`: fix creation of temporary socket files following
+:gh:`137335` (`PR#148578 <https://github.com/python/cpython/pull/148578>`__
+and corresponding backports). Patch by Bénédikt Tran.