You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
mitogen/ansible_mitogen/connection.py

1031 lines
34 KiB
Python

# Copyright 2017, David Wilson
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# 1. Redistributions of source code must retain the above copyright notice,
# this list of conditions and the following disclaimer.
#
# 2. Redistributions in binary form must reproduce the above copyright notice,
# this list of conditions and the following disclaimer in the documentation
# and/or other materials provided with the distribution.
#
# 3. Neither the name of the copyright holder nor the names of its contributors
# may be used to endorse or promote products derived from this software without
# specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
# ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
from __future__ import absolute_import
from __future__ import unicode_literals
import errno
import logging
import os
import pprint
import random
import stat
import sys
import time
7 years ago
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
import jinja2.runtime
import ansible.constants as C
import ansible.errors
7 years ago
import ansible.plugins.connection
import ansible.utils.shlex
import mitogen.core
import mitogen.fork
import mitogen.unix
import mitogen.utils
7 years ago
import ansible_mitogen.parsing
import ansible_mitogen.process
import ansible_mitogen.services
import ansible_mitogen.target
import ansible_mitogen.transport_config
7 years ago
LOG = logging.getLogger(__name__)
def optional_int(value):
"""
Convert `value` to an integer if it is not :data:`None`, otherwise return
:data:`None`.
"""
try:
return int(value)
except (TypeError, ValueError):
return None
def convert_bool(obj):
if isinstance(obj, bool):
return obj
if str(obj).lower() in ('no', 'false', '0'):
return False
if str(obj).lower() not in ('yes', 'true', '1'):
raise ansible.errors.AnsibleConnectionFailure(
'expected yes/no/true/false/0/1, got %r' % (obj,)
)
return True
def default(value, default):
"""
Return `default` is `value` is :data:`None`, otherwise return `value`.
"""
if value is None:
return default
return value
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
def _connect_local(spec):
"""
Return ContextService arguments for a local connection.
"""
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
return {
'method': 'local',
'kwargs': {
'python_path': spec.python_path(),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
}
}
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
def _connect_ssh(spec):
"""
Return ContextService arguments for an SSH connection.
"""
if C.HOST_KEY_CHECKING:
check_host_keys = 'enforce'
else:
check_host_keys = 'ignore'
# #334: tilde-expand private_key_file to avoid implementation difference
# between Python and OpenSSH.
private_key_file = spec.private_key_file()
if private_key_file is not None:
private_key_file = os.path.expanduser(private_key_file)
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
return {
'method': 'ssh',
'kwargs': {
'check_host_keys': check_host_keys,
'hostname': spec.remote_addr(),
'username': spec.remote_user(),
'compression': convert_bool(
default(spec.mitogen_ssh_compression(), True)
),
'password': spec.password(),
'port': spec.port(),
'python_path': spec.python_path(),
'identity_file': private_key_file,
'identities_only': False,
'ssh_path': spec.ssh_executable(),
'connect_timeout': spec.ansible_ssh_timeout(),
'ssh_args': spec.ssh_args(),
'ssh_debug_level': spec.mitogen_ssh_debug_level(),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
}
}
def _connect_docker(spec):
"""
Return ContextService arguments for a Docker connection.
"""
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
return {
'method': 'docker',
'kwargs': {
'username': spec.remote_user(),
'container': spec.remote_addr(),
'python_path': spec.python_path(),
'connect_timeout': spec.ansible_ssh_timeout() or spec.timeout(),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
}
}
def _connect_kubectl(spec):
"""
Return ContextService arguments for a Kubernetes connection.
"""
return {
'method': 'kubectl',
'kwargs': {
'pod': spec.remote_addr(),
'python_path': spec.python_path(),
'connect_timeout': spec.ansible_ssh_timeout() or spec.timeout(),
'kubectl_path': spec.mitogen_kubectl_path(),
'kubectl_args': spec.extra_args(),
}
}
def _connect_jail(spec):
"""
Return ContextService arguments for a FreeBSD jail connection.
"""
return {
'method': 'jail',
'kwargs': {
'username': spec.remote_user(),
'container': spec.remote_addr(),
'python_path': spec.python_path(),
'connect_timeout': spec.ansible_ssh_timeout() or spec.timeout(),
}
}
def _connect_lxc(spec):
"""
Return ContextService arguments for an LXC Classic container connection.
"""
return {
'method': 'lxc',
'kwargs': {
'container': spec.remote_addr(),
'python_path': spec.python_path(),
'lxc_attach_path': spec.mitogen_lxc_attach_path(),
'connect_timeout': spec.ansible_ssh_timeout() or spec.timeout(),
}
}
def _connect_lxd(spec):
"""
Return ContextService arguments for an LXD container connection.
"""
return {
'method': 'lxd',
'kwargs': {
'container': spec.remote_addr(),
'python_path': spec.python_path(),
'lxc_path': spec.mitogen_lxc_path(),
'connect_timeout': spec.ansible_ssh_timeout() or spec.timeout(),
}
}
def _connect_machinectl(spec):
"""
Return ContextService arguments for a machinectl connection.
"""
return _connect_setns(spec, kind='machinectl')
def _connect_setns(spec, kind=None):
"""
Return ContextService arguments for a mitogen_setns connection.
"""
return {
'method': 'setns',
'kwargs': {
'container': spec.remote_addr(),
'username': spec.remote_user(),
'python_path': spec.python_path(),
'kind': kind or spec.mitogen_kind(),
'docker_path': spec.mitogen_docker_path(),
'lxc_path': spec.mitogen_lxc_path(),
'lxc_info_path': spec.mitogen_lxc_info_path(),
'machinectl_path': spec.mitogen_machinectl_path(),
}
}
def _connect_su(spec):
"""
Return ContextService arguments for su as a become method.
"""
return {
'method': 'su',
'enable_lru': True,
'kwargs': {
'username': spec.become_user(),
'password': spec.become_pass(),
'python_path': spec.python_path(),
'su_path': spec.become_exe(),
'connect_timeout': spec.timeout(),
}
}
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
def _connect_sudo(spec):
"""
Return ContextService arguments for sudo as a become method.
"""
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
return {
'method': 'sudo',
'enable_lru': True,
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
'kwargs': {
'username': spec.become_user(),
'password': spec.become_pass(),
'python_path': spec.python_path(),
'sudo_path': spec.become_exe(),
'connect_timeout': spec.timeout(),
'sudo_args': spec.sudo_args(),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
}
}
def _connect_doas(spec):
"""
Return ContextService arguments for doas as a become method.
"""
return {
'method': 'doas',
'enable_lru': True,
'kwargs': {
'username': spec.become_user(),
'password': spec.become_pass(),
'python_path': spec.python_path(),
'doas_path': spec.become_exe(),
'connect_timeout': spec.timeout(),
}
}
def _connect_mitogen_su(spec):
"""
Return ContextService arguments for su as a first class connection.
"""
return {
'method': 'su',
'kwargs': {
'username': spec.remote_user(),
'password': spec.password(),
'python_path': spec.python_path(),
'su_path': spec.become_exe(),
'connect_timeout': spec.timeout(),
}
}
def _connect_mitogen_sudo(spec):
"""
Return ContextService arguments for sudo as a first class connection.
"""
return {
'method': 'sudo',
'kwargs': {
'username': spec.remote_user(),
'password': spec.password(),
'python_path': spec.python_path(),
'sudo_path': spec.become_exe(),
'connect_timeout': spec.timeout(),
'sudo_args': spec.sudo_args(),
}
}
def _connect_mitogen_doas(spec):
"""
Return ContextService arguments for doas as a first class connection.
"""
return {
'method': 'doas',
'kwargs': {
'username': spec.remote_user(),
'password': spec.password(),
'python_path': spec.python_path(),
'doas_path': spec.become_exe(),
'connect_timeout': spec.timeout(),
}
}
#: Mapping of connection method names to functions invoked as `func(spec)`
#: generating ContextService keyword arguments matching a connection
#: specification.
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
CONNECTION_METHOD = {
'docker': _connect_docker,
'kubectl': _connect_kubectl,
'jail': _connect_jail,
'local': _connect_local,
'lxc': _connect_lxc,
'lxd': _connect_lxd,
'machinectl': _connect_machinectl,
'setns': _connect_setns,
'ssh': _connect_ssh,
'su': _connect_su,
'sudo': _connect_sudo,
'doas': _connect_doas,
'mitogen_su': _connect_mitogen_su,
'mitogen_sudo': _connect_mitogen_sudo,
'mitogen_doas': _connect_mitogen_doas,
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
}
class Broker(mitogen.master.Broker):
"""
WorkerProcess maintains at most 2 file descriptors, therefore does not need
the exuberant syscall expense of EpollPoller, so override it and restore
the poll() poller.
"""
poller_class = mitogen.core.Poller
class CallChain(mitogen.parent.CallChain):
"""
Extend :class:`mitogen.parent.CallChain` to additionally cause the
associated :class:`Connection` to be reset if a ChannelError occurs.
This only catches failures that occur while a call is pending, it is a
stop-gap until a more general method is available to notice connection in
every situation.
"""
call_aborted_msg = (
'Mitogen was disconnected from the remote environment while a call '
'was in-progress. If you feel this is in error, please file a bug. '
'Original error was: %s'
)
def __init__(self, connection, context, pipelined=False):
super(CallChain, self).__init__(context, pipelined)
#: The connection to reset on CallError.
self._connection = connection
def _rethrow(self, recv):
try:
return recv.get().unpickle()
except mitogen.core.ChannelError as e:
self._connection.reset()
raise ansible.errors.AnsibleConnectionFailure(
self.call_aborted_msg % (e,)
)
def call(self, func, *args, **kwargs):
"""
Like :meth:`mitogen.parent.CallChain.call`, but log timings.
"""
t0 = time.time()
try:
recv = self.call_async(func, *args, **kwargs)
return self._rethrow(recv)
finally:
LOG.debug('Call took %d ms: %r', 1000 * (time.time() - t0),
mitogen.parent.CallSpec(func, args, kwargs))
7 years ago
class Connection(ansible.plugins.connection.ConnectionBase):
#: mitogen.master.Broker for this worker.
broker = None
#: mitogen.master.Router for this worker.
router = None
#: mitogen.parent.Context representing the parent Context, which is
issue #186: rework async/forked tasks again. The controller must know the ID of the forked child in order to propagate dependencies to it, so forking+starting the module run cannot happen entirely on the target, without some additional mechanism to wait-and-repropagate the deps as they arrive on the target. Rework things so that init_child() also handles starting the fork parent, and returns it along with the context's home directory in a single round trip. Now master knows the identity of the fork parent, it can directly create fork children and call run_module_async() in them. This necessitates 2 roundtrips to start an asynchronous task. This whole thing sucks and entirely needs simplified, but for now things almost work, so keeping it. connection.py: * Expect ContextService to return the entire dict return value of init_child(). Store the fork_contxt from the return value. planner.py: * Rework Planner to store the invocation as an instance attribute, to simplify method calls. * Add Planner.get_push_files() and Planner.get_module_deps(). * Add _propagate_deps() which takes a Planner and ensures the deps it describes are sent to a (non forked or forked) context. * Move async task logic out of target.py and into invoke() / _invoke_*(). process.py: * Services no longer need references to each other. planner.py handles sending module deps with one extra RPC. services.py: * Return "init_child_result" key instead of simple "home_dir" key. * Get rid of dep propagation from ModuleDepService, it lives in planner.py now. target.py: * Get rid of async task start logic, lives in planner.py now.
7 years ago
#: presently always the connection multiplexer process.
parent = None
#: mitogen.parent.Context for the target account on the target, possibly
#: reached via become.
7 years ago
context = None
#: Context for the login account on the target. This is always the login
#: account, even when become=True.
login_context = None
#: Only sudo, su, and doas are supported for now.
become_methods = ['sudo', 'su', 'doas']
#: Dict containing init_child() return value as recorded at startup by
#: ContextService. Contains:
#:
#: fork_context: Context connected to the fork parent : process in the
#: target account.
#: home_dir: Target context's home directory.
#: good_temp_dir: A writeable directory where new temporary directories
#: can be created.
init_child_result = None
issue #186: rework async/forked tasks again. The controller must know the ID of the forked child in order to propagate dependencies to it, so forking+starting the module run cannot happen entirely on the target, without some additional mechanism to wait-and-repropagate the deps as they arrive on the target. Rework things so that init_child() also handles starting the fork parent, and returns it along with the context's home directory in a single round trip. Now master knows the identity of the fork parent, it can directly create fork children and call run_module_async() in them. This necessitates 2 roundtrips to start an asynchronous task. This whole thing sucks and entirely needs simplified, but for now things almost work, so keeping it. connection.py: * Expect ContextService to return the entire dict return value of init_child(). Store the fork_contxt from the return value. planner.py: * Rework Planner to store the invocation as an instance attribute, to simplify method calls. * Add Planner.get_push_files() and Planner.get_module_deps(). * Add _propagate_deps() which takes a Planner and ensures the deps it describes are sent to a (non forked or forked) context. * Move async task logic out of target.py and into invoke() / _invoke_*(). process.py: * Services no longer need references to each other. planner.py handles sending module deps with one extra RPC. services.py: * Return "init_child_result" key instead of simple "home_dir" key. * Get rid of dep propagation from ModuleDepService, it lives in planner.py now. target.py: * Get rid of async task start logic, lives in planner.py now.
7 years ago
#: A :class:`mitogen.parent.CallChain` for calls made to the target
#: account, to ensure subsequent calls fail with the original exception if
#: pipelined directory creation or file transfer fails.
chain = None
#
# Note: any of the attributes below may be :data:`None` if the connection
# plugin was constructed directly by a non-cooperative action, such as in
# the case of the synchronize module.
#
#: Set to the host name as it appears in inventory by on_action_run().
inventory_hostname = None
#: Set to task_vars by on_action_run().
_task_vars = None
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
#: Set to 'hostvars' by on_action_run()
host_vars = None
#: Set by on_action_run()
delegate_to_hostname = None
#: Set to '_loader.get_basedir()' by on_action_run(). Used by mitogen_local
#: to change the working directory to that of the current playbook,
#: matching vanilla Ansible behaviour.
loader_basedir = None
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
def __init__(self, play_context, new_stdin, **kwargs):
assert ansible_mitogen.process.MuxProcess.unix_listener_path, (
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
'Mitogen connection types may only be instantiated '
'while the "mitogen" strategy is active.'
)
super(Connection, self).__init__(play_context, new_stdin)
def __del__(self):
"""
Ansible cannot be trusted to always call close() e.g. the synchronize
action constructs a local connection like this. So provide a destructor
in the hopes of catching these cases.
"""
# https://github.com/dw/mitogen/issues/140
self.close()
def on_action_run(self, task_vars, delegate_to_hostname, loader_basedir):
"""
Invoked by ActionModuleMixin to indicate a new task is about to start
executing. We use the opportunity to grab relevant bits from the
task-specific data.
:param dict task_vars:
Task variable dictionary.
:param str delegate_to_hostname:
:data:`None`, or the template-expanded inventory hostname this task
is being delegated to. A similar variable exists on PlayContext
when ``delegate_to:`` is active, however it is unexpanded.
:param str loader_basedir:
Loader base directory; see :attr:`loader_basedir`.
"""
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
self.inventory_hostname = task_vars['inventory_hostname']
self._task_vars = task_vars
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
self.host_vars = task_vars['hostvars']
self.delegate_to_hostname = delegate_to_hostname
self.loader_basedir = loader_basedir
self._mitogen_reset(mode='put')
def get_task_var(self, key, default=None):
"""
Fetch the value of a task variable related to connection configuration,
or, if delegate_to is active, fetch the same variable via HostVars for
the delegated-to machine.
When running with delegate_to, Ansible tasks have variables associated
with the original machine, not the delegated-to machine, therefore it
does not make sense to extract connection-related configuration for the
delegated-to machine from them.
"""
if self._task_vars:
if self.delegate_to_hostname is None:
if key in self._task_vars:
return self._task_vars[key]
else:
delegated_vars = self._task_vars['ansible_delegated_vars']
if self.delegate_to_hostname in delegated_vars:
task_vars = delegated_vars[self.delegate_to_hostname]
if key in task_vars:
return task_vars[key]
return default
@property
def homedir(self):
self._connect()
return self.init_child_result['home_dir']
7 years ago
@property
def connected(self):
return self.context is not None
7 years ago
def _spec_from_via(self, proxied_inventory_name, via_spec):
"""
Produce a dict connection specifiction given a string `via_spec`, of
the form `[[become_method:]become_user@]inventory_hostname`.
"""
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
become_user, _, inventory_name = via_spec.rpartition('@')
become_method, _, become_user = become_user.rpartition(':')
# must use __contains__ to avoid a TypeError for a missing host on
# Ansible 2.3.
if self.host_vars is None or inventory_name not in self.host_vars:
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
raise ansible.errors.AnsibleConnectionFailure(
self.unknown_via_msg % (
via_spec,
proxied_inventory_name,
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
)
)
via_vars = self.host_vars[inventory_name]
return ansible_mitogen.transport_config.MitogenViaSpec(
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
inventory_name=inventory_name,
play_context=self._play_context,
host_vars=dict(via_vars), # TODO: make it lazy
become_method=become_method or None,
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
become_user=become_user or None,
)
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
unknown_via_msg = 'mitogen_via=%s of %s specifies an unknown hostname'
via_cycle_msg = 'mitogen_via=%s of %s creates a cycle (%s)'
def _stack_from_spec(self, spec, stack=(), seen_names=()):
"""
Return a tuple of ContextService parameter dictionaries corresponding
to the connection described by `spec`, and any connection referenced by
its `mitogen_via` or `become` fields. Each element is a dict of the
form::
{
# Optional. If present and `True`, this hop is elegible for
# interpreter recycling.
"enable_lru": True,
# mitogen.master.Router method name.
"method": "ssh",
# mitogen.master.Router method kwargs.
"kwargs": {
"hostname": "..."
}
}
:param ansible_mitogen.transport_config.Spec spec:
Connection specification.
:param tuple stack:
Stack elements from parent call (used for recursion).
:param tuple seen_names:
Inventory hostnames from parent call (cycle detection).
:returns:
Tuple `(stack, seen_names)`.
"""
if spec.inventory_name() in seen_names:
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
raise ansible.errors.AnsibleConnectionFailure(
self.via_cycle_msg % (
spec.mitogen_via(),
spec.inventory_name(),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
' -> '.join(reversed(
seen_names + (spec.inventory_name(),)
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
)),
)
)
if spec.mitogen_via():
stack = self._stack_from_spec(
self._spec_from_via(spec.inventory_name(), spec.mitogen_via()),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
stack=stack,
seen_names=seen_names + (spec.inventory_name(),),
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
)
stack += (CONNECTION_METHOD[spec.transport()](spec),)
if spec.become() and ((spec.become_user() != spec.remote_user()) or
C.BECOME_ALLOW_SAME_USER):
stack += (CONNECTION_METHOD[spec.become_method()](spec),)
return stack
def _connect_broker(self):
"""
Establish a reference to the Broker, Router and parent context used for
connections.
"""
if not self.broker:
self.broker = mitogen.master.Broker()
self.router, self.parent = mitogen.unix.connect(
path=ansible_mitogen.process.MuxProcess.unix_listener_path,
broker=self.broker,
)
def _build_stack(self):
"""
Construct a list of dictionaries representing the connection
configuration between the controller and the target. This is
additionally used by the integration tests "mitogen_get_stack" action
to fetch the would-be connection configuration.
"""
return self._stack_from_spec(
ansible_mitogen.transport_config.PlayContextSpec(
connection=self,
play_context=self._play_context,
transport=self.transport,
inventory_name=self.inventory_hostname,
)
)
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
def _connect_stack(self, stack):
"""
Pass `stack` to ContextService, requesting a copy of the context object
representing the last tuple element. If no connection exists yet,
ContextService will recursively establish it before returning it or
throwing an error.
See :meth:`ansible_mitogen.services.ContextService.get` docstring for
description of the returned dictionary.
"""
try:
dct = self.parent.call_service(
service_name='ansible_mitogen.services.ContextService',
method_name='get',
stack=mitogen.utils.cast(list(stack)),
)
except mitogen.core.CallError:
LOG.warning('Connection failed; stack configuration was:\n%s',
pprint.pformat(stack))
raise
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
if dct['msg']:
if dct['method_name'] in self.become_methods:
raise ansible.errors.AnsibleModuleError(dct['msg'])
raise ansible.errors.AnsibleConnectionFailure(dct['msg'])
ansible: connection delegation v1 This implements the first edition of Connection Delegation, where delegating connection establishment is initially single-threaded. ansible_mitogen/strategy.py: ansible_mitogen/plugins/connection/*: Begin splitting connection.Connection into subclasses, exposing them directly as "mitogen_ssh", "mitogen_local", etc. connection types. This is far from removing strategy.py, but it's a tiny start. ansible_mitogen/connection.py: * config_from_play_context() and config_from_host_vars() build up a huge dictionary containing either more or less PlayContext contents, or our best attempt at reconstructing a host's connection config from its hostvars, where that config is not the current WorkerProcess target. They both produce the same format with the same keys, allowing remaining code to have a single input format. These dicts contain fields named after how Ansible refers to them, e.g. "sudo_exe". * _config_from_via() parses a basic connection specification like "username@inventory_name" into one of the aforementioned dicts. * _stack_from_config() produces a list of dicts describing the order in which (Mitogen) connections should be established, such that each element is proxied via= the previous element. The dicts produced by this function use Mitogen keyword arguments, the former di. These dicts contain fields named after how Mitogen refers to them, e.g. "sudo_path". * Pass the stack to ContextService, which is responsible for actual setup of the full chain. ansible_mitogen/services.py: Teach get() to walk the supplied stack, establishing each connection in turn, creating refounts for it before continuing. TODO: refcounting is broken in a variety of cases.
7 years ago
self.context = dct['context']
self.chain = CallChain(self, self.context, pipelined=True)
if self._play_context.become:
self.login_context = dct['via']
else:
self.login_context = self.context
self.init_child_result = dct['init_child_result']
def get_good_temp_dir(self):
"""
Return the 'good temporary directory' as discovered by
:func:`ansible_mitogen.target.init_child` immediately after
ContextService constructed the target context.
"""
self._connect()
return self.init_child_result['good_temp_dir']
def _generate_tmp_path(self):
return os.path.join(
self.get_good_temp_dir(),
'ansible_mitogen_action_%016x' % (
random.getrandbits(8*8),
)
)
def _make_tmp_path(self):
assert getattr(self._shell, 'tmpdir', None) is None
self._shell.tmpdir = self._generate_tmp_path()
LOG.debug('Temporary directory: %r', self._shell.tmpdir)
self.get_chain().call_no_reply(os.mkdir, self._shell.tmpdir)
return self._shell.tmpdir
def _reset_tmp_path(self):
"""
Called by _mitogen_reset(); ask the remote context to delete any
temporary directory created for the action. CallChain is not used here
to ensure exception is logged by the context on failure, since the
CallChain itself is about to be destructed.
"""
if getattr(self._shell, 'tmpdir', None) is not None:
self.context.call_no_reply(
ansible_mitogen.target.prune_tree,
self._shell.tmpdir,
)
self._shell.tmpdir = None
def _connect(self):
"""
Establish a connection to the master process's UNIX listener socket,
constructing a mitogen.master.Router to communicate with the master,
and a mitogen.parent.Context to represent it.
Depending on the original transport we should emulate, trigger one of
the _connect_*() service calls defined above to cause the master
process to establish the real connection on our behalf, or return a
reference to the existing one.
"""
if self.connected:
return
self._connect_broker()
stack = self._build_stack()
self._connect_stack(stack)
def _mitogen_reset(self, mode):
"""
Forget everything we know about the connected context. This function
cannot be called _reset() since that name is used as a public API by
Ansible 2.4 wait_for_connection plug-in.
:param str mode:
Name of ContextService method to use to discard the context, either
'put' or 'reset'.
"""
if not self.context:
return
self._reset_tmp_path()
self.chain.reset()
self.parent.call_service(
service_name='ansible_mitogen.services.ContextService',
method_name=mode,
context=self.context
)
self.context = None
self.login_context = None
self.init_child_result = None
self.chain = None
def _shutdown_broker(self):
"""
Shutdown the broker thread during :meth:`close` or :meth:`reset`.
"""
if self.broker:
self.broker.shutdown()
self.broker.join()
self.broker = None
self.router = None
# #420: Ansible executes "meta" actions in the top-level process,
# meaning "reset_connection" will cause :class:`mitogen.core.Latch`
# FDs to be cached and erroneously shared by children on subsequent
# WorkerProcess forks. To handle that, call on_fork() to ensure any
# shared state is discarded.
# #490: only attempt to clean up when it's known that some
# resources exist to cleanup, otherwise later __del__ double-call
# to close() due to GC at random moment may obliterate an unrelated
# Connection's resources.
mitogen.fork.on_fork()
def close(self):
"""
Arrange for the mitogen.master.Router running in the worker to
gracefully shut down, and wait for shutdown to complete. Safe to call
multiple times.
"""
self._mitogen_reset(mode='put')
self._shutdown_broker()
def _reset_find_task_vars(self):
"""
Monsterous hack: since "meta: reset_connection" does not run from an
action, we cannot capture task variables via :meth:`on_action_run`.
Instead walk the parent frames searching for the `all_vars` local from
StrategyBase._execute_meta(). If this fails, just leave task_vars
unset, likely causing a subtly wrong configuration to be selected.
"""
frame = sys._getframe()
while frame and not self._task_vars:
self._task_vars = frame.f_locals.get('all_vars')
frame = frame.f_back
reset_compat_msg = (
'Mitogen only supports "reset_connection" on Ansible 2.5.6 or later'
)
def reset(self):
"""
Explicitly terminate the connection to the remote host. This discards
any local state we hold for the connection, returns the Connection to
the 'disconnected' state, and informs ContextService the connection is
bad somehow, and should be shut down and discarded.
"""
if self._task_vars is None:
self._reset_find_task_vars()
if self._play_context.remote_addr is None:
# <2.5.6 incorrectly populate PlayContext for reset_connection
# https://github.com/ansible/ansible/issues/27520
raise ansible.errors.AnsibleConnectionFailure(
self.reset_compat_msg
)
self._connect()
self._mitogen_reset(mode='reset')
self._shutdown_broker()
# Compatibility with Ansible 2.4 wait_for_connection plug-in.
_reset = reset
def get_chain(self, use_login=False, use_fork=False):
"""
Return the :class:`mitogen.parent.CallChain` to use for executing
function calls.
:param bool use_login:
If :data:`True`, always return the chain for the login account
rather than any active become user.
:param bool use_fork:
If :data:`True`, return the chain for the fork parent.
:returns mitogen.parent.CallChain:
"""
self._connect()
if use_login:
return self.login_context.default_call_chain
# See FORK_SUPPORTED comments in target.py.
if use_fork and self.init_child_result['fork_context'] is not None:
return self.init_child_result['fork_context'].default_call_chain
return self.chain
7 years ago
def spawn_isolated_child(self):
issue #186: rework async/forked tasks again. The controller must know the ID of the forked child in order to propagate dependencies to it, so forking+starting the module run cannot happen entirely on the target, without some additional mechanism to wait-and-repropagate the deps as they arrive on the target. Rework things so that init_child() also handles starting the fork parent, and returns it along with the context's home directory in a single round trip. Now master knows the identity of the fork parent, it can directly create fork children and call run_module_async() in them. This necessitates 2 roundtrips to start an asynchronous task. This whole thing sucks and entirely needs simplified, but for now things almost work, so keeping it. connection.py: * Expect ContextService to return the entire dict return value of init_child(). Store the fork_contxt from the return value. planner.py: * Rework Planner to store the invocation as an instance attribute, to simplify method calls. * Add Planner.get_push_files() and Planner.get_module_deps(). * Add _propagate_deps() which takes a Planner and ensures the deps it describes are sent to a (non forked or forked) context. * Move async task logic out of target.py and into invoke() / _invoke_*(). process.py: * Services no longer need references to each other. planner.py handles sending module deps with one extra RPC. services.py: * Return "init_child_result" key instead of simple "home_dir" key. * Get rid of dep propagation from ModuleDepService, it lives in planner.py now. target.py: * Get rid of async task start logic, lives in planner.py now.
7 years ago
"""
Fork or launch a new child off the target context.
issue #186: rework async/forked tasks again. The controller must know the ID of the forked child in order to propagate dependencies to it, so forking+starting the module run cannot happen entirely on the target, without some additional mechanism to wait-and-repropagate the deps as they arrive on the target. Rework things so that init_child() also handles starting the fork parent, and returns it along with the context's home directory in a single round trip. Now master knows the identity of the fork parent, it can directly create fork children and call run_module_async() in them. This necessitates 2 roundtrips to start an asynchronous task. This whole thing sucks and entirely needs simplified, but for now things almost work, so keeping it. connection.py: * Expect ContextService to return the entire dict return value of init_child(). Store the fork_contxt from the return value. planner.py: * Rework Planner to store the invocation as an instance attribute, to simplify method calls. * Add Planner.get_push_files() and Planner.get_module_deps(). * Add _propagate_deps() which takes a Planner and ensures the deps it describes are sent to a (non forked or forked) context. * Move async task logic out of target.py and into invoke() / _invoke_*(). process.py: * Services no longer need references to each other. planner.py handles sending module deps with one extra RPC. services.py: * Return "init_child_result" key instead of simple "home_dir" key. * Get rid of dep propagation from ModuleDepService, it lives in planner.py now. target.py: * Get rid of async task start logic, lives in planner.py now.
7 years ago
:returns:
mitogen.core.Context of the new child.
"""
return self.get_chain(use_fork=True).call(
ansible_mitogen.target.spawn_isolated_child
)
issue #186: rework async/forked tasks again. The controller must know the ID of the forked child in order to propagate dependencies to it, so forking+starting the module run cannot happen entirely on the target, without some additional mechanism to wait-and-repropagate the deps as they arrive on the target. Rework things so that init_child() also handles starting the fork parent, and returns it along with the context's home directory in a single round trip. Now master knows the identity of the fork parent, it can directly create fork children and call run_module_async() in them. This necessitates 2 roundtrips to start an asynchronous task. This whole thing sucks and entirely needs simplified, but for now things almost work, so keeping it. connection.py: * Expect ContextService to return the entire dict return value of init_child(). Store the fork_contxt from the return value. planner.py: * Rework Planner to store the invocation as an instance attribute, to simplify method calls. * Add Planner.get_push_files() and Planner.get_module_deps(). * Add _propagate_deps() which takes a Planner and ensures the deps it describes are sent to a (non forked or forked) context. * Move async task logic out of target.py and into invoke() / _invoke_*(). process.py: * Services no longer need references to each other. planner.py handles sending module deps with one extra RPC. services.py: * Return "init_child_result" key instead of simple "home_dir" key. * Get rid of dep propagation from ModuleDepService, it lives in planner.py now. target.py: * Get rid of async task start logic, lives in planner.py now.
7 years ago
def get_extra_args(self):
"""
Overridden by connections/mitogen_kubectl.py to a list of additional
arguments for the command.
"""
# TODO: maybe use this for SSH too.
return []
def get_default_cwd(self):
"""
Overridden by connections/mitogen_local.py to emulate behaviour of CWD
being fixed to that of ActionBase._loader.get_basedir().
"""
return None
def get_default_env(self):
"""
Overridden by connections/mitogen_local.py to emulate behaviour of
WorkProcess environment inherited from WorkerProcess.
"""
return None
def exec_command(self, cmd, in_data='', sudoable=True, mitogen_chdir=None):
"""
Implement exec_command() by calling the corresponding
ansible_mitogen.target function in the target.
:param str cmd:
Shell command to execute.
:param bytes in_data:
Data to supply on ``stdin`` of the process.
:returns:
(return code, stdout bytes, stderr bytes)
"""
emulate_tty = (not in_data and sudoable)
rc, stdout, stderr = self.get_chain().call(
ansible_mitogen.target.exec_command,
cmd=mitogen.utils.cast(cmd),
in_data=mitogen.utils.cast(in_data),
chdir=mitogen_chdir or self.get_default_cwd(),
emulate_tty=emulate_tty,
)
stderr += b'Shared connection to %s closed.%s' % (
self._play_context.remote_addr.encode(),
(b'\r\n' if emulate_tty else b'\n'),
)
return rc, stdout, stderr
7 years ago
def fetch_file(self, in_path, out_path):
"""
Implement fetch_file() by calling the corresponding
ansible_mitogen.target function in the target.
:param str in_path:
Remote filesystem path to read.
:param str out_path:
Local filesystem path to write.
"""
output = self.get_chain().call(
ansible_mitogen.target.read_path,
mitogen.utils.cast(in_path),
)
ansible_mitogen.target.write_path(out_path, output)
7 years ago
def put_data(self, out_path, data, mode=None, utimes=None):
"""
Implement put_file() by caling the corresponding ansible_mitogen.target
function in the target, transferring small files inline. This is
pipelined and will return immediately; failed transfers are reported as
exceptions in subsequent functon calls.
:param str out_path:
Remote filesystem path to write.
:param byte data:
File contents to put.
"""
self.get_chain().call_no_reply(
ansible_mitogen.target.write_path,
mitogen.utils.cast(out_path),
mitogen.core.Blob(data),
mode=mode,
utimes=utimes,
)
#: Maximum size of a small file before switching to streaming
#: transfer. This should really be the same as
#: mitogen.services.FileService.IO_SIZE, however the message format has
#: slightly more overhead, so just randomly subtract 4KiB.
SMALL_FILE_LIMIT = mitogen.core.CHUNK_SIZE - 4096
def _throw_io_error(self, e, path):
if e.args[0] == errno.ENOENT:
s = 'file or module does not exist: ' + path
raise ansible.errors.AnsibleFileNotFound(s)
7 years ago
def put_file(self, in_path, out_path):
"""
Implement put_file() by streamily transferring the file via
FileService.
:param str in_path:
Local filesystem path to read.
:param str out_path:
Remote filesystem path to write.
"""
try:
st = os.stat(in_path)
except OSError as e:
self._throw_io_error(e, in_path)
raise
if not stat.S_ISREG(st.st_mode):
raise IOError('%r is not a regular file.' % (in_path,))
# If the file is sufficiently small, just ship it in the argument list
# rather than introducing an extra RTT for the child to request it from
# FileService.
if st.st_size <= self.SMALL_FILE_LIMIT:
try:
fp = open(in_path, 'rb')
try:
s = fp.read(self.SMALL_FILE_LIMIT + 1)
finally:
fp.close()
except OSError:
self._throw_io_error(e, in_path)
raise
# Ensure did not grow during read.
if len(s) == st.st_size:
return self.put_data(out_path, s, mode=st.st_mode,
utimes=(st.st_atime, st.st_mtime))
self._connect()
self.parent.call_service(
service_name='mitogen.service.FileService',
method_name='register',
path=mitogen.utils.cast(in_path)
)
# For now this must remain synchronous, as the action plug-in may have
# passed us a temporary file to transfer. A future FileService could
# maintain an LRU list of open file descriptors to keep the temporary
# file alive, but that requires more work.
self.get_chain().call(
ansible_mitogen.target.transfer_file,
context=self.parent,
in_path=in_path,
out_path=out_path
)