Add some extra information to an async failure message to reflect the
actually timeout value of the failure.
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
* fix tagged implicit gathering
- None is not tHe default anymore, its an empty list
* Update changelogs/fragments/fix_taggged_gather.yml
Co-Authored-By: bcoca <bcoca@users.noreply.github.com>
* Added basic equivalent to PowerShell modules
* changes based on latest review
* Added tests
* ignore sanity test due to how tests are set up
* Changes to work with PSCore
* Added documentation and change updated more modules
* Add some speed optimisations to AddType
* fix some issues in the doc changes
* doc changes
When python is compiled in debug mode, or certain command line flags are
passed, python issues helpful warnings to let users know of files opened
by not closed.
This can be fairly spammy on an ansible run. This change reduces the
number of such warnings by a factor of 10.
* win_exec: refactor PS exec runner
* more changes for PSCore compatibility
* made some changes based on the recent review
* split up module exec scripts for smaller payload
* removed C# module support to focus on just error msg improvement
* cleaned up c# test classifier code
* Try to intuit proper plugins to send to ansible-connection
* Move sub-plugins to init so that vars will be populated in executor
* Fix connection unit tests
* win async: use async_dir for the async results file directory
* tried to unify POSIX and PowerShell async implementations of async_dir
* fix sanity issue
* win_script: add support for become and centralise exec wrapper builder
* satisfying the pep8 gods
* do not scan for module dependencies when running as a script
The AnsiballZ optimization which only uses one pyton interpreter
currently monkeypatches the arguments into a global argument in module_utils
so we need to always include basic.py. In the future we should fix this
so that it monkeypatches its own file. That way we won't need to always
include basic.py
Extends `module_defaults` by adding a prefix to defaults `group/` which denotes a builtin list of modules. Initial groups are: `group/aws`, `group/azure`, and `group/gcp`
* Don't use copy.deepcopy in high workload areas, use deepishcopy. ci_complete
* Add tests
* Add changelog fragment
* rename to naive_deepcopy and add extra docs
* Rename to module_response_deepcopy and move to vars/clean
* restore task arg splatting
* reverts #41804
* supersedes #41295
* fixes#42192
* after lots of discussion amongst the core team, we decided to preserve this feature, clarify the runtime warnings/docs, and prioritize a path toward fixing the underlying behavior that causes this feature to be insecure (un-namespaced facts).
* update faq text
note that warning is disabled when inject_facts_as_vars is
* wordsmithing FAQ entry
* Only template values in vars_prompt rather than all vars
This allows the use of variables in vars_prompt fields but allows variables entered in the prompt to affect play vars rather than throwing an undefined error.
Only post validate if there was a vars_prompt
* Add tests for vars_prompt
* Change how data is sent to the persistent connection socket.
We can't rely on readline(), so send the size of the data first. We can
then read that many bytes from the stream on the recieving end.
* Set pty to noncanonical mode before sending
* Now that we send data length, we don't need a sentinel anymore
* Copy socket changes to persistent, too
* Use os.write instead of fdopen()ing and using that.
* Follow pickle with sha1sum of pickle
* Swap order of vars and init being passed to ansible-connection
Now that we don't need to worry about python-2.4 and 2.5, we can make
some improvements to the way AnsiballZ handles modules.
* Change AnsiballZ wrapper to use import to invoke the module
We need the module to think of itself as a script because it could be
coded as:
main()
or as:
if __name__ == '__main__':
main()
Or even as:
if __name__ == '__main__':
random_function_name()
A script will invoke all of those. Prior to this change, we invoked
a second Python interpreter on the module so that it really was
a script. However, this means that we have to run python twice (once
for the AnsiballZ wrapper and once for the module). This change makes
the module think that it is a script (because __name__ in the module ==
'__main__') but it's actually being invoked by us importing the module
code.
There's three ways we've come up to do this.
* The most elegant is to use zipimporter and tell the import mechanism
that the module being loaded is __main__:
* 5959f11c9d/lib/ansible/executor/module_common.py (L175)
* zipimporter is nice because we do not have to extract the module from
the zip file and save it to the disk when we do that. The import
machinery does it all for us.
* The drawback is that modules do not have a __file__ which points
to a real file when they do this. Modules could be using __file__
to for a variety of reasons, most of those probably have
replacements (the most common one is to find a writable directory
for temporary files. AnsibleModule.tmpdir should be used instead)
We can monkeypatch __file__ in fom AnsibleModule initialization
but that's kind of gross. There's no way I can see to do this
from the wrapper.
* Next, there's imp.load_module():
* https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151
* imp has the nice property of allowing us to set __name__ to
__main__ without changing the name of the file itself
* We also don't have to do anything special to set __file__ for
backwards compatibility (although the reason for that is the
drawback):
* Its drawback is that it requires the file to exist on disk so we
have to explicitly extract it from the zipfile and save it to
a temporary file
* The last choice is to use exec to execute the module:
* https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175
* The code we would have to maintain for this looks pretty clean.
In the wrapper we create a ModuleType, set __file__ on it, read
the module's contents in from the zip file and then exec it.
* Drawbacks: We still have to explicitly extract the file's contents
from the zip archive instead of letting python's import mechanism
handle it.
* Exec also has hidden performance issues and breaks certain
assumptions that modules could be making about their own code:
http://lucumr.pocoo.org/2011/2/1/exec-in-python/
Our plan is to use imp.load_module() for now, deprecate the use of
__file__ in modules, and switch to zipimport once the deprecation
period for __file__ is over (without monkeypatching a fake __file__ in
via AnsibleModule).
* Rename the name of the AnsiBallZ wrapped module
This makes it obvious that the wrapped module isn't the module file that
we distribute. It's part of trying to mitigate the fact that the module
is now named __main)).py in tracebacks.
* Shield all wrapper symbols inside of a function
With the new import code, all symbols in the wrapper become visible in
the module. To mitigate the chance of collisions, move most symbols
into a toplevel function. The only symbols left in the global namespace
are now _ANSIBALLZ_WRAPPER and _ansiballz_main.
revised porting guide entry
Integrate code coverage collection into AnsiballZ.
ci_coverage
ci_complete
* fix minor issues with debug and item labels
- no more `item=None`, we always have a label now
- debug should only show expected information, either msg= or the var in var=
- also fixed method name, deprecated misleading _get_item
Also fixes a discovered bug in block rescue detection related to inserting
the ansible_failed_{result|task} variables when the rescue is in a nested
block.
* I seem to have forgotten the back half of tests
* Set http timeout from persistent_command_timeout
* Tweak URL generation and provide URL on error
* Push var_options to connection process
* Don't wait forever if coming from persistent
* Don't send the entire contents of variables to ansible-connection
* WIP Pull persistent connection parameters via get_option
* Fix pep8
* Add use_persistent_connection setting to paramiko_ssh plugin
* Add vars section to persistent_command_timeout setting and prevail provider values over config manager
* Use persistent_command_timeout on network_cli instead of timeout
* Fix unit tests
If we don't call loader to get network_cli, then _load_name is never
set and we get KeyError.
* Pull persistent_command_timeout via config manager for ios connection local
* Pull persistent_command_timeout via config manager on connection local
* configurable list of facts modules
- allow for args dict for specific modules
- add way to pass parameters
- avoid facts poluting test
- move to 'facts gathered' flag
- add 'gathering' setting tests
* task_executor.py: Raise 'conditional exception' in case of 'include_*'
Fixes#33632
Signed-off-by: Patrick Ringl <patrick_@freenet.de>
* Re-organize tests, add static include test with undefined var
* Remove unnecessary conditional
this change further enables the ansible-vcr project to collect fixtures per host in
situations where the host context is totally lost due to delegation.
* Keep some original pre-delegate values in connection options
* More concisely reparent, ensuring we don't go too shallow or too deep in this process. Fixes#38357
* More explicit reparenting, with a short circuit for a common case
* We need new_block to have a parent, otherwise we lose context with this approach
* Remove duplicate parent assignment
* Change callers of Block.copy to not use exclude_parent=True, when including the parent, exclude tasks
* ignore version when deciding callback loading
The code already defaulted to load the callback if the properties are not present
there was no need for us to also check the version
fixes#38270
* fix error msg on set optoins to use correct name
* tolerate windows line endings when loading windows module utils. Helpful for old custom windows modules.
* add test modules to demonstrate win line ending module load behaviour.
* attempt to fix sanity check failures
* pep8 fix
* explict skip of test modules from shebang check (core modules must still have expected unix style line endings)
* switch to rstrip() following core team meeting feedback
* module_common: set required parameter templar
Fix the following error (related to b455901):
$ ./hacking/test-module -m ./lib/ansible/modules/system/ping.py -I ansible_python_interpreter=/usr/bin/python
Traceback (most recent call last):
File "./hacking/test-module", line 268, in <module>
main()
File "./hacking/test-module", line 249, in main
(modfile, modname, module_style) = boilerplate_module(options.module_path, options.module_args, interpreters, options.check, options.filename)
File "./hacking/test-module", line 152, in boilerplate_module
task_vars=task_vars
File "ansible/lib/ansible/executor/module_common.py", line 910, in modify_module
environment=environment)
File "ansible/lib/ansible/executor/module_common.py", line 736, in _find_module_utils
shebang, interpreter = _get_shebang(u'/usr/bin/python', task_vars, templar)
File "ansible/lib/ansible/executor/module_common.py", line 452, in _get_shebang
interpreter = templar.template(task_vars[interpreter_config].strip())
AttributeError: 'NoneType' object has no attribute 'template'
* module_common.modify_module: templar is required
include_vars and set_fact are already updating hostvars in strategy
no need to 're add again' with lower priority the same data.
fixes#37535, mostly by avoiding reprocessing and 'cleaning'
* Check that connection error msg are not unsafe
* Connection error messages are unsafe: wrap them
For example, in case of error, docker connection plugin returns exception
message containing Go template. These messages weren't tagged as unsafe
and were consequently rendered:
The conditional check 'result is failed' failed. The error was:
{
'msg': u'Docker version check ([\'/usr/bin/docker\', \'version\', \'--format\', "\'{{.Server.Version}}\'"]) failed: Got permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Get http://%2Fvar%2Frun%2Fdocker.sock/v1.35/version: dial unix /var/run/docker.sock: connect: permission denied\n',
'failed': True
}:
template error while templating string: unexpected '.'.
String: Docker version check (['/usr/bin/docker', 'version', '--format', "'{{.Server.Version}}'"]) failed: Got permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Get http://%2Fvar%2Frun%2Fdocker.sock/v1.35/version: dial unix /var/run/docker.sock: connect: permission denied
* Fix types when evaluating interpreter. Fixes#36536
* Rename variables that contain bytes to b_*
* Get rid of to_text() and to_bytes() calls that do nothing (because the
data is already the proper type)
* fix loop_control templating
- properly template ALL properites for loop_control
- loop_control inherits from base, it should not, but needs validate/attribute functionality
fixes#24719
* Fix debug logs failing with persistent connection
Fixes#33047
* As debug logs are written on stdout, it interrupts
the communication between ansible-connection(background)
process and main process. To avoid this add a string similar
to exactly identify the response string.
* Remove unwanted code in ansible-connection
* Fix review comments
* Fix spurious log emitted on ansible-connection stdout issue
* ansible-connection which runs as a background process sends a
json string (contains response received from remote device)
to foreground ansible-playbook process over stdout.
* If in case debug flag is enabled the connection_loader api
invoked from ansible-connection `ssh = connection_loader.get('ssh', class_only=True)`
results in emitting debug logs on stdout. This spurious log
interfere with the actual response and results in failure while
reading json string in ansible-playbook process
* To avoid this save stdout of ansible-connection and redirect it string
buffer to accumulate all the logs emitted by core API's
* Add these logs in `result['messages']` which is send a json string after reinstating saved stdout
* Remove unwanted code in ansible-connection
* Fix review comment
* allow shells to have per host options, remote_tmp
added language to shell
removed module lang setting from general as plugins have it now
use get to avoid bad powershell plugin
more resilient tmp discovery, fall back to `pwd`
add shell to docs
fixed options for when frags are only options
added shell set ops in t_e and fixed option frags
normalize tmp dir usag4e
- pass tmpdir/tmp/temp options as env var to commands, making it default for tempfile
- adjusted ansiballz tmpdir
- default local tempfile usage to the configured local tmp
- set env temp in action
add options to powershell
shift temporary to internal envvar/params
ensure tempdir is set if we pass var
ensure basic and url use expected tempdir
ensure localhost uses local tmp
give /var/tmp priority, less perms issues
more consistent tempfile mgmt for ansiballz
made async_dir configurable
better action handling, allow for finally rm tmp
fixed tmp issue and no more tempdir in ballz
hostvarize world readable and admin users
always set shell tempdir
added comment to discourage use of exception/flow control
* Mostly revert expand_user as it's not quite working.
This was an additional feature anyhow.
Kept the use of pwd as a fallback but moved it to a second ssh
connection. This is not optimal but getting that to work in a single
ssh connection was part of the problem holding this up.
(cherry picked from commit 395b714120522f15e4c90a346f5e8e8d79213aca)
* fixed script and other action plugins
ensure tmpdir deletion
allow for connections that don't support new options (legacy, 3rd party)
fixed tests
* Cache tasks as they are queued instead of en masse
This also moves the task caching from the PlayIterator to the
StrategyBase class, where it makes more sense (and makes it easier
to not have to change the strategy class methods leading to an API
change).
Fixes#31673
* Cleaning up unit tests due to 502ca780
* Fix persistent command timeout handling
We were using play context timeout on ansible-connect to set
the persistent command timeout handler. Thus, we were ignoring
the persistent_command_timeout setting.
Moreover, even by changing that on ansible-connection, were again
overriding it on cliconf send_command, since in a same process we can
just set a single alarm and cliconf send_command alarm setup is executed
after ansible-connection alarm setup.
* Remove alarm setting on cliconf send_command
The alarm is set regardless before it is executed by ansible-connection.
Setting an alarm again, overrides/disables the previous ones as a single
process can just have a single alarm set.
* Move the setting of persistent command timeout to network_cli
We do also use ansible-connection for connection local, so if a
user provides a timeout via provider that would be ignored if we
set the value on ansible-connection.
Moving that logic to network_cli plugin constructor makes both
connections to work.
* Remove debug statements
* Set the persistent command timeout on task_executor
We can't set the timeout on ansible-connection nor network_cli,
otherwise tasks using provider timeout won't work.
* Provide a way to explicitly invoke the debugger with in the debug strategy
* Merge the debugger strategy into StrategyBase
* Fix some logic, pin to a single result
* Make redo also continue
* Make sure that if the debug closure doesn't need to process the result, that we still return it
* Fix failing tests for the strategy
* Clean up messages from debugger and exit code to match bin/ansible
* Move the FieldAttribute higher, to apply at different levels
* make debugger a string, expand logic
* Better host state rollbacks
* More explicit debugger prompt
* ENABLE_TASK_DEBUGGER should be boolean, and better docs
* No bare except, add pprint, alias h, vars to task_vars
* _validate_debugger can ignore non-string, that can be caught later
* Address issue if there were no previous tasks/state, and use the correct key
* Update docs for changes to the debugger
* Guard against a stat going negative through use of decrement
* Add a few notes about using the debugger on the free strategy
* Add changelog entry for task debugger
* Add a few versionadded indicators and a note about vars -> task_vars
* Add integration test for missing attempts in until with failed/changed_when
* Extend tests to validate additional known behaviors of do-until loops
* Inject attempts into result earlier
The short-circuiting of include_role in TaskExecutor was using the _task object
in the result dictionary. This causes the playbook class to be serialized, which
is something we no longer do due to speed/size problems with nested objects.
Simply removing this should fix a lot of memory and speed problems with include_role.
* Add parent pid to persistent connection socket path hash
Fixes#33192
* Add parent pid in persistent connection socket path hash
to avoid using same socket path for multiple simultaneous
connection to same remote host.
* Ensure unique persistent socket path for each ansible-playbook run
* Fix CI failures
* Adding aliases for field attributes and renaming async attribute
As of Python 3.7, the use of async raises an error, whereas before the use
of the reserved word was ignored. This adds an alias field for field attrs
so that both async and async_val (interally) work. This allows us to be
backwards-compatible with 3rd party plugins that may still reference Task.async,
but for the core engine to work on Py3.7+.
* Remove files fixed for 'async' usage from the python 3.7 skip list
* included inventory and callback in new config
allow inventory to be configurable
updated connection options settings
also updated winrm to work with new configs
removed now obsolete set_host_overrides
added notes for future bcoca, current one is just punting, it's future's problem
updated docs per feedback
added remove group/host methods to inv data
moved fact cache from data to constructed
cleaner/better options
fix when vars are added
extended ignore list to config dicts
updated paramiko connection docs
removed options from base that paramiko already handles
left the look option as it is used by other plugin types
resolve delegation
updated cache doc options
fixed test_script
better fragment merge for options
fixed proxy command
restore ini for proxy
normalized options
moved pipelining to class
updates for host_key_checking
restructured mixins
* fix typo
* implements jsonrpc message passing for ansible-connection
* implements more generic mechanism for persistent connections
* starts persistent connection in task_executor if enabled and supported
* supports using network_cli as top level connection plugin
* enhances logging for persistent connection to stdout
* Update action plugins
* Fix Python3 RPC
* Fix Junos bytes<-->str issues
* supports using netconf as top level connection plugin
* Error message when running netconf on an unsupported platform
* Update tests
* Fix `authorize: yes` for `connection: local`
* Handle potentially JSON data in terminal
* Add clarifying detail if possible on ConnectionError
currently it is doing only from the 'active' hosts in the batch which means
the percentage goes up as hosts fail instead of staying the same.
added debug info for max fail
fixes#32255
* windows: add #AnsibleRequires to set whether a module requires module or a specific version
* fix up pep8 issues
* changed psversion to use the actual ps Requires -Version syntax
* missed the check on #Requires -Version
* fix #Requires module extensions
Abhijeet Kasurde
Martin Krizek
Jordan Borean
Brian Coca
Matt Martz
Eitan Adler
Matt Clay
Nathaniel Case
Brian Coca
Toshio Kuratomi
Ryan Brown
Matt Davis
Sam Doran
James Cammarata
Ricardo Carrillo Cruz
Andrew Gaffney
Mickaël Guérin
jctanner
jhawkesworth
Pilou
Dag Wieers
Ganesh Nalawade
Sam Doran