Connection plugins network_cli and netconf (#32521)

* implements jsonrpc message passing for ansible-connection

* implements more generic mechanism for persistent connections
* starts persistent connection in task_executor if enabled and supported
* supports using network_cli as top level connection plugin
* enhances logging for persistent connection to stdout

* Update action plugins

* Fix Python3 RPC

* Fix Junos bytes<-->str issues

* supports using netconf as top level connection plugin

* Error message when running netconf on an unsupported platform
* Update tests

* Fix `authorize: yes` for `connection: local`

* Handle potentially JSON data in terminal

* Add clarifying detail if possible on ConnectionError
This commit is contained in:
Nathaniel Case 2017-11-09 15:04:40 -05:00 committed by GitHub
commit 9c0275a879
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
26 changed files with 722 additions and 798 deletions

View file

@ -91,33 +91,14 @@ def fail_if_missing(module, found, service, msg=''):
module.fail_json(msg='Could not find the requested service %s: %s' % (service, msg))
def daemonize(module, cmd):
def fork_process():
'''
Execute a command while detaching as a daemon, returns rc, stdout, and stderr.
:arg module: is an AnsibleModule object, used for it's utility methods
:arg cmd: is a list or string representing the command and options to run
This is complex because daemonization is hard for people.
What we do is daemonize a part of this module, the daemon runs the command,
picks up the return code and output, and returns it to the main process.
This function performs the double fork process to detach from the
parent process and execute.
'''
pid = os.fork()
# init some vars
chunk = 4096 # FIXME: pass in as arg?
errors = 'surrogate_or_strict'
# start it!
try:
pipe = os.pipe()
pid = os.fork()
except OSError:
module.fail_json(msg="Error while attempting to fork: %s", exception=traceback.format_exc())
# we don't do any locking as this should be a unique module/process
if pid == 0:
os.close(pipe[0])
# Set stdin/stdout/stderr to /dev/null
fd = os.open(os.devnull, os.O_RDWR)
@ -140,7 +121,7 @@ def daemonize(module, cmd):
# get new process session and detach
sid = os.setsid()
if sid == -1:
module.fail_json(msg="Unable to detach session while daemonizing")
raise Exception("Unable to detach session while daemonizing")
# avoid possible problems with cwd being removed
os.chdir("/")
@ -149,6 +130,38 @@ def daemonize(module, cmd):
if pid > 0:
os._exit(0)
return pid
def daemonize(module, cmd):
'''
Execute a command while detaching as a daemon, returns rc, stdout, and stderr.
:arg module: is an AnsibleModule object, used for it's utility methods
:arg cmd: is a list or string representing the command and options to run
This is complex because daemonization is hard for people.
What we do is daemonize a part of this module, the daemon runs the command,
picks up the return code and output, and returns it to the main process.
'''
# init some vars
chunk = 4096 # FIXME: pass in as arg?
errors = 'surrogate_or_strict'
# start it!
try:
pipe = os.pipe()
pid = fork_process()
except OSError:
module.fail_json(msg="Error while attempting to fork: %s", exception=traceback.format_exc())
except Exception as exc:
module.fail_json(msg=to_text(exc), exception=traceback.format_exc())
# we don't do any locking as this should be a unique module/process
if pid == 0:
os.close(pipe[0])
# if command is string deal with py2 vs py3 conversions for shlex
if not isinstance(cmd, list):
if PY2: