Python + Ubuntu Linux + nohup error: [1]+ Exit

Question

I have a python script, practice_one.py, that I’d like it to run forever in Ubuntu Linux, and has the following:

while True:
   #    Code

And I attempted nohup python practice_one.py & but got the message nohup: ignoring input and appending output to ‘nohup.out’.

Then when I hit the enter key, the outputs another message: [1]+ Exit nohup python practice_one.py

How come it just exits automatically? What could I be doing wrong?

EDIT

Attempted: nohup python practice_one.py </dev/null &>/dev/null &

And get [1] 61122 then when I press enter I got [1]+ Exit nohup python practice_one.py </dev/null &>/dev/null &

It used to work but now exits.. What could be the issue now?

Answer 1

I usually face this in three cases:

1. When an exception in the while loop occurs.

2. When somewhere in the code I have pdb.set_trace() which is a debugging breakpoint.

3. When somewhere in the code, input from user is requested as in raw_input("Please enter something: ")

Check your code to exclude these three causes.

Don't worry about the first message, it is just informing you that nohup is in work.

EDIT

4. Before running the script with nohup make sure that the script itself works fine (This would actually be the first thing to try).

Answer 2

I suggest that you'd better use double fork magic to run your long-running program as a daemon instead of nohup. It is easy to debug and your program doesn't need external tools like nohup.

daemon.py

import sys
import os

# double fork magic
def daemonize():
    os.umask(0)

    try:
        pid = os.fork()
    except OSError:
        sys.exit(0)
    if pid > 0:
        sys.exit(0)
    os.setsid()

    try:
        pid = os.fork()
    except OSError:
        sys.exit(0)
    if pid > 0:
        os._exit(0)

    os.chdir('/')

    import resource     # Resource usage information.
    maxfd = resource.getrlimit(resource.RLIMIT_NOFILE)[1]
    if (maxfd == resource.RLIM_INFINITY):
        maxfd = 1024

    # Iterate through and close all file descriptors.
    for fd in range(0, maxfd):
        try:
            os.close(fd)
        except OSError:   # ERROR, fd wasn't open to begin with (ignored)
            pass

    fd = os.open(os.devnull, os.O_RDWR)
    os.dup2(fd, sys.stdin.fileno())
    os.dup2(fd, sys.stdout.fileno())
    os.dup2(fd, sys.stderr.fileno())

test.py

from daemon import daemonize
import time


def test():
    while True:
        time.sleep(10)

if __name__ == '__main__':
    daemonize() # you could comment this line before you make sure your program run as you expect
    test()

Now, using python test.py to make it run as a daemon and you could use ps aux | grep test.py to check it.