The ohai and facter modules use /usr/bin/logger to log the fact that
they have been invoked.  I added 'import os' to the ping module
so that it could have the same syslog statements as the other modules.
I separated the condensed:
    shlex.split(open(argfile, 'r').read())
into two separate statements similar to the other modules.

Applying callback model to runner, and using that in playbooks, so output can be more immediate in playbooks.

(Runner still does not use callbacks for default output)

Fix async to use the new argfiles method (wrapping brain around rock, really must write module development guide)

happens only on needed hosts, allowing some hosts to fail and drop out of the running.

Parameters:
  jid=X
  mode=status|cleanup (default status)

status = returns results from the job
cleanup = deletes the job file, should also kill the job if still running (TODO)