community.general/lib/ansible/plugins/strategy/free.py
Matt Martz 445ff39f94
Become plugins (#50991)
* [WIP] become plugins

Move from hardcoded method to plugins for ease of use, expansion and overrides
  - load into connection as it is going to be the main consumer
  - play_context will also use to keep backwards compat API
  - ensure shell is used to construct commands when needed
  - migrate settings remove from base config in favor of plugin specific configs
  - cleanup ansible-doc
  - add become plugin docs
  - remove deprecated sudo/su code and keywords
  - adjust become options for cli
  - set plugin options from context
  - ensure config defs are avaialbe before instance
  - refactored getting the shell plugin, fixed tests
     - changed into regex as they were string matching, which does not work with random string generation
     - explicitly set flags for play context tests
 - moved plugin loading up front
 - now loads for basedir also
 - allow pyc/o for non m modules
 - fixes to tests and some plugins
 - migrate to play objects fro play_context
 - simiplify gathering
 -  added utf8 headers
 - moved option setting
 - add fail msg to dzdo
 - use tuple for multiple options on fail/missing
 - fix relative plugin paths
 - shift from play context to play
 - all tasks already inherit this from play directly
 - remove obsolete 'set play'
 - correct environment handling
 - add wrap_exe option to pfexec
 - fix runas to noop
 - fixed setting play context
 - added password configs
 - removed required false
 - remove from doc building till they are ready

future development:
  - deal with 'enable' and 'runas' which are not 'command wrappers' but 'state flags' and currently hardcoded in diff subsystems

* cleanup

  remove callers to removed func
  removed --sudo cli doc refs
  remove runas become_exe
  ensure keyerorr on plugin
  also fix backwards compat, missing method is attributeerror, not ansible error
  get remote_user consistently
  ignore missing system_tmpdirs on plugin load
  correct config precedence
  add deprecation
  fix networking imports
  backwards compat for plugins using BECOME_METHODS

* Port become_plugins to context.CLIARGS

This is a work in progress:
* Stop passing options around everywhere as we can use context.CLIARGS
  instead

* Refactor make_become_commands as asked for by alikins

* Typo in comment fix

* Stop loading values from the cli in more than one place

Both play and play_context were saving default values from the cli
arguments directly.  This changes things so that the default values are
loaded into the play and then play_context takes them from there.

* Rename BECOME_PLUGIN_PATH to DEFAULT_BECOME_PLUGIN_PATH

As alikins said, all other plugin paths are named
DEFAULT_plugintype_PLUGIN_PATH.  If we're going to rename these, that
should be done all at one time rather than piecemeal.

* One to throw away

This is a set of hacks to get setting FieldAttribute defaults to command
line args to work.  It's not fully done yet.

After talking it over with sivel and jimi-c this should be done by
fixing FieldAttributeBase and _get_parent_attribute() calls to do the
right thing when there is a non-None default.

What we want to be able to do ideally is something like this:

class Base(FieldAttributeBase):
    _check_mode = FieldAttribute([..] default=lambda: context.CLIARGS['check'])

class Play(Base):
    # lambda so that we have a chance to parse the command line args
    # before we get here.  In the future we might be able to restructure
    # this so that the cli parsing code runs before these classes are
    # defined.

class Task(Base):
    pass

And still have a playbook like this function:

---
- hosts:
  tasks:
  - command: whoami
    check_mode: True

(The check_mode test that is added as a separate commit in this PR will
let you test variations on this case).

There's a few separate reasons that the code doesn't let us do this or
a non-ugly workaround for this as written right now.  The fix that
jimi-c, sivel, and I talked about may let us do this or it may still
require a workaround (but less ugly) (having one class that has the
FieldAttributes with default values and one class that inherits from
that but just overrides the FieldAttributes which now have defaults)

* Revert "One to throw away"

This reverts commit 23aa883cbed11429ef1be2a2d0ed18f83a3b8064.

* Set FieldAttr defaults directly from CLIARGS

* Remove dead code

* Move timeout directly to PlayContext, it's never needed on Play

* just for backwards compat, add a static version of BECOME_METHODS to constants

* Make the become attr on the connection public, since it's used outside of the connection

* Logic fix

* Nuke connection testing if it supports specific become methods

* Remove unused vars

* Address rebase issues

* Fix path encoding issue

* Remove unused import

* Various cleanups

* Restore network_cli check in _low_level_execute_command

* type improvements for cliargs_deferred_get and swap shallowcopy to default to False

* minor cleanups

* Allow the su plugin to work, since it doesn't define a prompt the same way

* Fix up ksu become plugin

* Only set prompt if build_become_command was called

* Add helper to assist connection plugins in knowing they need to wait for a prompt

* Fix tests and code expectations

* Doc updates

* Various additional minor cleanups

* Make doas functional

* Don't change connection signature, load become plugin from TaskExecutor

* Remove unused imports

* Add comment about setting the become plugin on the playcontext

* Fix up tests for recent changes

* Support 'Password:' natively for the doas plugin

* Make default prompts raw

* wording cleanups. ci_complete

* Remove unrelated changes

* Address spelling mistake

* Restore removed test, and udpate to use new functionality

* Add changelog fragment

* Don't hard fail in set_attributes_from_cli on missing CLI keys

* Remove unrelated change to loader

* Remove internal deprecated FieldAttributes now

* Emit deprecation warnings now
2019-02-11 11:27:44 -06:00

257 lines
13 KiB
Python

# (c) 2012-2014, Michael DeHaan <michael.dehaan@gmail.com>
#
# This file is part of Ansible
#
# Ansible is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# Ansible is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Ansible. If not, see <http://www.gnu.org/licenses/>.
# Make coding more python3-ish
from __future__ import (absolute_import, division, print_function)
__metaclass__ = type
DOCUMENTATION = '''
strategy: free
short_description: Executes tasks without waiting for all hosts
description:
- Task execution is as fast as possible per batch as defined by C(serial) (default all).
Ansible will not wait for other hosts to finish the current task before queuing more tasks for other hosts.
All hosts are still attempted for the current task, but it prevents blocking new tasks for hosts that have already finished.
- With the free strategy, unlike the default linear strategy, a host that is slow or stuck on a specific task
won't hold up the rest of the hosts and tasks.
version_added: "2.0"
author: Ansible Core Team
'''
import time
from ansible import constants as C
from ansible.errors import AnsibleError
from ansible.playbook.included_file import IncludedFile
from ansible.plugins.loader import action_loader
from ansible.plugins.strategy import StrategyBase
from ansible.template import Templar
from ansible.module_utils._text import to_text
from ansible.utils.display import Display
display = Display()
class StrategyModule(StrategyBase):
def _filter_notified_hosts(self, notified_hosts):
'''
Filter notified hosts accordingly to strategy
'''
# We act only on hosts that are ready to flush handlers
return [host for host in notified_hosts
if host in self._flushed_hosts and self._flushed_hosts[host]]
def __init__(self, tqm):
super(StrategyModule, self).__init__(tqm)
self._host_pinned = False
def run(self, iterator, play_context):
'''
The "free" strategy is a bit more complex, in that it allows tasks to
be sent to hosts as quickly as they can be processed. This means that
some hosts may finish very quickly if run tasks result in little or no
work being done versus other systems.
The algorithm used here also tries to be more "fair" when iterating
through hosts by remembering the last host in the list to be given a task
and starting the search from there as opposed to the top of the hosts
list again, which would end up favoring hosts near the beginning of the
list.
'''
# the last host to be given a task
last_host = 0
result = self._tqm.RUN_OK
# start with all workers being counted as being free
workers_free = len(self._workers)
work_to_do = True
while work_to_do and not self._tqm._terminated:
hosts_left = self.get_hosts_left(iterator)
if len(hosts_left) == 0:
self._tqm.send_callback('v2_playbook_on_no_hosts_remaining')
result = False
break
work_to_do = False # assume we have no more work to do
starting_host = last_host # save current position so we know when we've looped back around and need to break
# try and find an unblocked host with a task to run
host_results = []
while True:
host = hosts_left[last_host]
display.debug("next free host: %s" % host)
host_name = host.get_name()
# peek at the next task for the host, to see if there's
# anything to do do for this host
(state, task) = iterator.get_next_task_for_host(host, peek=True)
display.debug("free host state: %s" % state, host=host_name)
display.debug("free host task: %s" % task, host=host_name)
if host_name not in self._tqm._unreachable_hosts and task:
# set the flag so the outer loop knows we've still found
# some work which needs to be done
work_to_do = True
display.debug("this host has work to do", host=host_name)
# check to see if this host is blocked (still executing a previous task)
if host_name not in self._blocked_hosts or not self._blocked_hosts[host_name]:
# pop the task, mark the host blocked, and queue it
self._blocked_hosts[host_name] = True
(state, task) = iterator.get_next_task_for_host(host)
try:
action = action_loader.get(task.action, class_only=True)
except KeyError:
# we don't care here, because the action may simply not have a
# corresponding action plugin
action = None
display.debug("getting variables", host=host_name)
task_vars = self._variable_manager.get_vars(play=iterator._play, host=host, task=task)
self.add_tqm_variables(task_vars, play=iterator._play)
templar = Templar(loader=self._loader, variables=task_vars)
display.debug("done getting variables", host=host_name)
try:
task.name = to_text(templar.template(task.name, fail_on_undefined=False), nonstring='empty')
display.debug("done templating", host=host_name)
except Exception:
# just ignore any errors during task name templating,
# we don't care if it just shows the raw name
display.debug("templating failed for some reason", host=host_name)
run_once = templar.template(task.run_once) or action and getattr(action, 'BYPASS_HOST_LOOP', False)
if run_once:
if action and getattr(action, 'BYPASS_HOST_LOOP', False):
raise AnsibleError("The '%s' module bypasses the host loop, which is currently not supported in the free strategy "
"and would instead execute for every host in the inventory list." % task.action, obj=task._ds)
else:
display.warning("Using run_once with the free strategy is not currently supported. This task will still be "
"executed for every host in the inventory list.")
# check to see if this task should be skipped, due to it being a member of a
# role which has already run (and whether that role allows duplicate execution)
if task._role and task._role.has_run(host):
# If there is no metadata, the default behavior is to not allow duplicates,
# if there is metadata, check to see if the allow_duplicates flag was set to true
if task._role._metadata is None or task._role._metadata and not task._role._metadata.allow_duplicates:
display.debug("'%s' skipped because role has already run" % task, host=host_name)
del self._blocked_hosts[host_name]
continue
if task.action == 'meta':
self._execute_meta(task, play_context, iterator, target_host=host)
self._blocked_hosts[host_name] = False
else:
# handle step if needed, skip meta actions as they are used internally
if not self._step or self._take_step(task, host_name):
if task.any_errors_fatal:
display.warning("Using any_errors_fatal with the free strategy is not supported, "
"as tasks are executed independently on each host")
self._tqm.send_callback('v2_playbook_on_task_start', task, is_conditional=False)
self._queue_task(host, task, task_vars, play_context)
# each task is counted as a worker being busy
workers_free -= 1
del task_vars
else:
display.debug("%s is blocked, skipping for now" % host_name)
# all workers have tasks to do (and the current host isn't done with the play).
# loop back to starting host and break out
if self._host_pinned and workers_free == 0 and work_to_do:
last_host = starting_host
break
# move on to the next host and make sure we
# haven't gone past the end of our hosts list
last_host += 1
if last_host > len(hosts_left) - 1:
last_host = 0
# if we've looped around back to the start, break out
if last_host == starting_host:
break
results = self._process_pending_results(iterator)
host_results.extend(results)
# each result is counted as a worker being free again
workers_free += len(results)
self.update_active_connections(results)
try:
included_files = IncludedFile.process_include_results(
host_results,
iterator=iterator,
loader=self._loader,
variable_manager=self._variable_manager
)
except AnsibleError as e:
return self._tqm.RUN_ERROR
if len(included_files) > 0:
all_blocks = dict((host, []) for host in hosts_left)
for included_file in included_files:
display.debug("collecting new blocks for %s" % included_file)
try:
if included_file._is_role:
new_ir = self._copy_included_file(included_file)
new_blocks, handler_blocks = new_ir.get_block_list(
play=iterator._play,
variable_manager=self._variable_manager,
loader=self._loader,
)
else:
new_blocks = self._load_included_file(included_file, iterator=iterator)
except AnsibleError as e:
for host in included_file._hosts:
iterator.mark_host_failed(host)
display.warning(to_text(e))
continue
for new_block in new_blocks:
task_vars = self._variable_manager.get_vars(play=iterator._play, task=new_block._parent)
final_block = new_block.filter_tagged_tasks(task_vars)
for host in hosts_left:
if host in included_file._hosts:
all_blocks[host].append(final_block)
display.debug("done collecting new blocks for %s" % included_file)
display.debug("adding all collected blocks from %d included file(s) to iterator" % len(included_files))
for host in hosts_left:
iterator.add_tasks(host, all_blocks[host])
display.debug("done adding collected blocks to iterator")
# pause briefly so we don't spin lock
time.sleep(C.DEFAULT_INTERNAL_POLL_INTERVAL)
# collect all the final results
results = self._wait_on_pending_results(iterator)
# run the base class run() method, which executes the cleanup function
# and runs any outstanding handlers which have been triggered
return super(StrategyModule, self).run(iterator, play_context, result)