basic local and remote communication

Comparing current working directories

A local popen gateway has the same working directory as the instantiatior:

>>> import execnet, os
>>> gw = execnet.makegateway()
>>> ch = gw.remote_exec("import os; channel.send(os.getcwd())")
>>> res = ch.receive()
>>> assert res == os.getcwd()

“ssh” gateways default to the login home directory.

Getting information from remote ssh account

Use simple execution to obtain information from remote environments:

>>> import execnet, os
>>> gw = execnet.makegateway("ssh=codespeak.net")
>>> channel = gw.remote_exec("""
...     import sys, os
...     channel.send((sys.platform, sys.version_info, os.getpid()))
... """)
>>> platform, version_info, remote_pid = channel.receive()
>>> platform
'linux2'
>>> version_info
(2, 4, 2, 'final', 0)

Avoid “inlined” source strings with remote_exec

You can pass a module object to remote_exec in which case its source code will be sent. No dependencies will be transferred so the module must be self-contained or only use modules that are installed on the “other” side. Module code can detect if it is running in a remote_exec situation by checking for the special __name__ attribute.

# content of a module remote1.py 

if __name__ == '__channelexec__':
    channel.send('initialization complete')

You can now remote-execute the module like this:

>>> import execnet, remote1
>>> gw = execnet.makegateway()
>>> ch = gw.remote_exec(remote1)
>>> print (ch.receive())
initialization complete

which will print the ‘initialization complete’ string.

a simple command loop pattern

If you want the remote side to serve a number of synchronous function calls into your module you can setup a serving loop and write a local protocol.

# contents of: remotecmd.py
def simple(arg):
    return arg + 1

if __name__ == '__channelexec__':
    for item in channel:
        channel.send(eval(item))

Then on the local side you can do:

>>> import execnet, remotecmd
>>> gw = execnet.makegateway()
>>> ch = gw.remote_exec(remotecmd)
>>> ch.send('simple(10)') # execute func-call remotely
>>> ch.receive()
11

Our remotecmd module starts up remote serving through the for item in channel loop which will terminate when the channel closes. It evaluates all incoming requests in the global name space and sends back the results.

Instantiate gateways through sockets

In cases where you do not have SSH-access to a machine you need to download a small version-independent standalone socketserver.py script to provide a remote bootstrapping-point. You do not need to install the execnet package remotely. Simply run the script like this:

python socketserver.py :8888   # bind to all IPs, port 8888

You can then instruct execnet on your local machine to bootstrap itself into the remote socket endpoint:

import execnet
gw = execnet.SocketGateway("TARGET-IP:8888")

That’s it, you can now use the gateway object just like a popen- or ssh-based one.