Manhole is in-process service that will accept unix domain socket connections and present the stacktraces for all threads and an interactive prompt. It can either work as a python daemon thread waiting for connections at all times or a signal handler (stopping your application and waiting for a connection).
Access to the socket is restricted to the application’s effective user id or root.
This is just like Twisted’s manhole. It’s simpler (no dependencies), it only runs on Unix domain sockets (in contrast to Twisted’s manhole which can run on telnet or ssh) and it integrates well with various types of applications.
pip install manhole
You can put this in your django settings, wsgi app file, some module that’s always imported early etc:
import manhole manhole.install() # this will start the daemon thread # and now you start your app, eg: server.serve_forever()
Now in a shell you can do either of these:
netcat -U /tmp/manhole-1234 socat - unix-connect:/tmp/manhole-1234 socat readline unix-connect:/tmp/manhole-1234
Socat with readline is best (history, editing etc). If your socat doesn’t have readme try this.
$ nc -U /tmp/manhole-1234 Python 2.7.3 (default, Apr 10 2013, 06:20:15) [GCC 4.6.3] on linux2 Type "help", "copyright", "credits" or "license" for more information. (InteractiveConsole) >>> dir() ['__builtins__', 'dump_stacktraces', 'os', 'socket', 'sys', 'traceback'] >>> print 'foobar' foobar
There’s a new experimental
manhole-cli bin since 1.1.0, that emulates
usage: manhole-cli [-h] [-t TIMEOUT] [-1 | -2 | -s SIGNAL] PID Connect to a manhole. positional arguments: PID A numerical process id, or a path in the form: /tmp/manhole-1234 optional arguments: -h, --help show this help message and exit -t TIMEOUT, --timeout TIMEOUT Timeout to use. Default: 1 seconds. -1, -USR1 Send USR1 (10) to the process before connecting. -2, -USR2 Send USR2 (12) to the process before connecting. -s SIGNAL, --signal SIGNAL Send the given SIGNAL to the process before connecting.
Uses unix domain sockets, only root or same effective user can connect.
Can run the connection in a thread or in a signal handler (see
Can start the thread listening for connections from a signal handler (see
Compatible with apps that fork, reinstalls the Manhole thread after fork – had to monkeypatch os.fork/os.forkpty for this.
Compatible with gevent and eventlet with some limitations – you need to either:
- Disable thread monkeypatching (eg:
Note: on eventlet you might need to setup the hub first to prevent circular import problems:
import eventlet eventlet.hubs.get_hub() # do this first eventlet.monkey_patch(thread=False)
The thread is compatible with apps that use signalfd (will mask all signals for the Manhole threads).
manhole.install( verbose=True, verbose_destination=2, patch_fork=True, activate_on=None, oneshot_on=None, sigmask=manhole.ALL_SIGNALS, socket_path=None, reinstall_delay=0.5, locals=None, strict=True, )
verbose– Set it to
Falseto squelch the logging.
verbose_destination– Destination for verbose messages. Set it to a file descriptor or handle. Default is unbuffered stderr (stderr
patch_fork– Set it to
Falseif you don’t want your
activate_on– Set to
"USR2"or some other signal name, or a number if you want the Manhole thread to start when this signal is sent. This is desirable in case you don’t want the thread active all the time.
thread– Set to
Trueto start the always-on ManholeThread. Default:
True. Automatically switched to
oneshot_on– Set to
"USR2"or some other signal name, or a number if you want the Manhole to listen for connection in the signal handler. This is desireable in case you don’t want threads at all.
sigmask– Will set the signal mask to the given list (using
signalfd.sigprocmask). No action is done if
signalfdis not importable. NOTE: This is done so that the Manhole thread doesn’t steal any signals; Normally that is fine because Python will force all the signal handling to be run in the main thread but signalfd doesn’t.
socket_path– Use a specific path for the unix domain socket (instead of
/tmp/manhole-<pid>). This disables
patch_forkas children cannot reuse the same path.
reinstall_delay– Delay the unix domain socket creation reinstall_delay seconds. This alleviates cleanup failures when using fork+exec patterns.
locals– Names to add to manhole interactive shell locals.
daemon_connection– The connection thread is daemonic (dies on app exit). Default:
redirect_stderr– Redirect output from stderr to manhole console. Default:
AlreadyInstalledwill be raised when attempting to install manhole twice. Default:
Environment variable installation
Manhole can be installed via the
PYTHONMANHOLE environment variable.
PYTHONMANHOLE='' python yourapp.py
Is equivalent to having this in
import manhole manhole.install()
Any extra text in the environment variable is passed to
PYTHONMANHOLE='onshot_on="USR2"' python yourapp.py
What happens when you actually connect to the socket
- Credentials are checked (if it’s same user or root)
sys.std*are redirected to the UDS
- Stacktraces for each thread are written to the UDS
- REPL is started so you can fiddle with the process
- Using threads and file handle (not raw file descriptor)
verbose_destinationcan cause deadlocks. See bug reports: PyPy and Python 3.4.
SIGTERM and socket cleanup
By default Python doesn’t call the
atexit callbacks with the default SIGTERM handling. This makes manhole leave stray socket files around. If this is undesirable you should install a custom SIGTERM handler so
atexit is properly invoked.
import signal import sys def handle_sigterm(signo, frame): sys.exit(128 + signo) # this will raise SystemExit and cause atexit to be called signal.signal(signal.SIGTERM, handle_sigterm)
Using Manhole with uWSGI
Because uWSGI overrides signal handling Manhole is a bit more tricky to setup. One way is to use “uWSGI signals” (not the POSIX signals) and have the workers check a file for the pid you want to open the Manhole in.
Stick something this in your WSGI application file:
from __future__ import print_function import sys import os import manhole stack_dump_file = '/tmp/manhole-pid' uwsgi_signal_number = 17 try: import uwsgi if not os.path.exists(stack_dump_file): open(stack_dump_file, 'w') def open_manhole(dummy_signum): with open(stack_dump_file, 'r') as fh: pid = fh.read().strip() if pid == str(os.getpid()): inst = manhole.install(strict=False, thread=False) inst.handle_oneshot(dummy_signum, dummy_signum) uwsgi.register_signal(uwsgi_signal_number, 'workers', open_manhole) uwsgi.add_file_monitor(uwsgi_signal_number, stack_dump_file) print("Listening for stack mahole requests via %r" % (stack_dump_file,), file=sys.stderr) except ImportError: print("Not running under uwsgi; unable to configure manhole trigger", file=sys.stderr) except IOError: print("IOError creating manhole trigger %r" % (stack_dump_file,), file=sys.stderr) # somewhere bellow you'd have something like from django.core.wsgi import get_wsgi_application application = get_wsgi_application() # or def application(environ, start_response): start_response('200 OK', [('Content-Type', 'text/plain'), ('Content-Length', '2')]) yield b'OK'
To open the Manhole just run echo 1234 > /tmp/manhole-pid and then manhole-cli 1234.
|OS:||Linux, OS X|
|Runtime:||Python 2.7, 3.4, 3.5, 3.6 or PyPy|
- Twisted’s manhole – it has colors and server-side history.
- wsgi-shell – spawns a thread.
- pyrasite – uses gdb to inject code.
- pydbattach – uses gdb to inject code.
- pystuck – very similar, uses rpyc for communication.
- pyringe – uses gdb to inject code, more reliable, but relies on dbg python builds unfortunatelly.
- pdb-clone – uses gdb to inject code, with a different strategy.