`get_options` is an undefined method from CallbackBase
As a result, remove the trailing `s` letter.
Signed-off-by: Daniel Andrei Minca <mandrei17@gmail.com>
* Add error handling for junos in case wrong connection type
Fixes#37990
If a junos module doesn't support given connection/transport type
return appropriate error message.
* Fix CI issues
* Fix review comment
* fixes issue when netconf would report ios is not supported
This change now will map ansible_network_os=ios to the correct netconf
plugin implementation. This will resolve an error where the netconf
connection plugin will report that ios is unsupported.
When the URI module returns complex JSON objects, the YAML callback
fails while trying to represent these objects. The problem arises
because the filter method returns an iterator in Python 3, rather than a
str object. Therefore, the str method expandtabs() is not available,
and the callback fails with the following error:
[WARNING]: Failure using method (v2_runner_on_failed) in callback plugin (<ansible.plugins.callback.yaml.CallbackModule object at 0x7f7c7ed8aa20>): 'filter' object has no attribute 'expandtabs'
Issue can be replicated by running this playbook:
- hosts: localhost
gather_facts: false
tasks:
- uri:
url: https://jsonplaceholder.typicode.com/posts
ansible-playbook tmp.yml -v
* 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
* Use arg_spec type for comparisons on default and choices
* Further improve type casting
* Make sure to capture output in more places
* Individually report invalid choices
* Update ignore.txt after resolving merge conflicts
Fixes#37208
If check_mode is enabled instead of committing th config need to
discard all the chnages to cnadidate db
In case of cli to discard changes issue `rollback 0` command
and for netconf execute `discard-changes` rpc call
This patch fixes up some English typos in the yaml inventory
plugin documentation.
* s/specifically/specific
* s/as/as an
Signed-off-by: Eric Brown <browne@vmware.com>
* template: Add integration tests for `lstrip_blocks'
Signed-off-by: Alex Tsitsimpis <alextsi@arrikto.com>
* template: Fix passing `trim_blocks' inline
Fix passing `trim_blocks' option to the template module as inline
argument. Previously passing the `trim_blocks' option inline instead of
using the YAML dictionary format resulted in it always being set to
`True', even if `trim_blocks=False' was used.
Signed-off-by: Alex Tsitsimpis <alextsi@arrikto.com>
* template: Add option to `lstrip_blocks'
Add option to set `lstrip_blocks' when using the template module to
render Jinja templates. The Jinja documentation suggests that
`trim_blocks' and `lstrip_blocks' is a great combination and the
template module already provides an option for `trim_blocks'.
Note that although `trim_blocks' in Ansible is enabled by default since
version 2.4, in order to avoid breaking things keep `lstrip_blocks'
disabled by default. Maybe in a future version it could be enabled by
default.
This seems to address issue #10725 in a more appropriate way than the
suggested.
Signed-off-by: Alex Tsitsimpis <alextsi@arrikto.com>
* template: Add integration tests for `trim_blocks'
Signed-off-by: Alex Tsitsimpis <alextsi@arrikto.com>
* template: Check Jinja2 support for `lstrip_blocks'
Since the `lstrip_blocks' option was added in Jinja2 version 2.7, raise
an exception when `lstrip_blocks' is set but Jinja2 does not support it.
Check support for `lstrip_blocks' option by checking `jinja2.defaults'
for `LSTRIP_BLOCKS' and do not use `jinja2.__version__' because the
latter is set to `unknown' in some cases, perhaps due to bug in
`pkg_resources' in Python 2.6.6.
Also update option description to state that Jinja2 version >=2.7 is
required.
Signed-off-by: Alex Tsitsimpis <alextsi@arrikto.com>
* Fix loading of filter and test plugins
Filter and test plugins are different than other plugins in that they
can have many plugins in a single file. Therefore they need to operate
a little differently. They need to have all of the potential files
returned. Then the caller takes care of passing those onto jinja2 in
order for jinja2 to make use of them.
This problem was (most recently) introduced with f921369445
This commit also restructures how we deduplicate plugins to take paths
into account. If we want to start scoping which set of modules are
loaded (due to roles, for instance) we'll need to hang on to the path
information.
* add integration test for override
* Fix style checks for bcoca code
* Implement jinja2 plugin loader as a subclass
Having a subclass allows us to customize the overriding of jinja
plugins. We can then move common parts of common code into the Loader.
* New generator inventory plugin
Allows construction of hosts and groups through cartesian
product of various group combinations
* Add generator plugin documentation
* Fix csvfile traceback on Python3
The csvfile lookup uses some custom iterators. These needed to be
ported to handle the python3 iterator protocol. In addition, the
csvfile module takes an iterator of byte strings in Python2 and an
iterator of text strings in Python3
Fixes#36808
* Remove default admin_distance and fix the idempotence thereof
Fixes#33290
* Fix tests and use yaml anchors to shorten tests
* Add test for undefined admin_distance
* Read config from `show run` if `show ip static route` fails
* Restore flags to ios.get_config & use get_config where appropriate
This commit fixes up the get_config method to match the minimum method
signature as defined by the base class. Without this patch, the
get_config method calls will fail in some cirumstances.
This PR includes:
- Further cleanup of BOTMETA.yml
- Cleanup of author github handles
- Use of proper booleans
- One-line license statement
- Copyright format statement
- Smaller cleanups
* If inventory file isn't able to be parsed by aws_ec2, return an empty dict instead of None
* Raise an AnsibleParserError instead
* remove extra lines
* aws_ec2 inventory plugin - fix path matching logic
Unified tmp accidentally removed the containing tmpdir from the list of
files to fix the permissions on when we're becoming a different
unprivileged user. This resulted in a visible bug for script but not
for patch. This is because patch also uploads the module to the same
temporary directory and the uploaded module also ends up calling
fixup_perms2() which includes the temporary directory. So by the time
patch needs to access the temporary patch file, the directory is
appropriately set.
script's breakage was visible because script does not upload a module
(it's akin to raw in this way). Therefore, we only call fixup_perms2()
once in script and so leaving out the tmpdir in script means that the
containing directory never has its permissions set appropriately.
Fixing both because it does not cause an extra round trip for patch so
any speedup would be minimal and it's better to fix the perms as close
as possible to where we know we need it. Otherwise, changes to
seemingly unrelated code later could end up breaking it.
Fixes#36398
send_command already performs the to_bytes safely on prompts (checking
for None). Without this check the literal 'None' became a subprompt trigger!
Fixes#35662
* Revert "commented out check tests (#35788)"
This reverts commit 3df2561405.
* `discard_changes()` now uses `exit discard`
instead of just `discard`. Also a slight tweak to avoid a second `exit` after
`exit discard`
* Exclude parent when copying included task to avoid memory issues. Fixes#35796
* Simplify implicit block squashing to pre-group, instead of post re-parenting
Implement the `v2_playbook_on_handler_task_start()' method in the json
callback plugin to correctly include handlers in the results.
This fixes a bug where the last task in the results returned by
json callback would be wrong if an unconditional handler was triggered,
since the result of that handler would overwrite the result of the last
task.
Signed-off-by: Alex Tsitsimpis <alextsi@arrikto.com>
* First bit of fixing temporary to have one source of truth
* Fix pep8
* Remove explicit make_tmp_path() in copy
The copy action plugin sets TRANSFER_FILES=True so it does not need to
set the temporary directory explicitly; the base class's run() method
will do that for us.
* Fix for calling a module's run when a shell has already created a temp path.
* Remember to inform the rest of the world when tempdir is removed
* New strategy for how to warn on passing tmp
Now we just warn when calling the parent class run() early. If the
module does a late call to the parent run() and doesn't make use of the
temporary directory, then we don't check for the possibility that the
user mistakenly is sending tmp in. If we truly deprecate this (rather
than ignoring it forever) then we might want to switch back to checking
for someone passing a value in as tmp.
* Remove tmp parameter from _execute_module as well
* Port all action plugins to not send tmp explicitly
This is now handled inside of _execute_module via the
_connection._shell.tempdir attribute.
Also update warnings and docs to tell people to set the attribute
instead of using _execute_module's tmp parameter.
* Always set local tempdir variable
* aws ssm parameter lookup test case - fails demonstrating no exception when parameter missing
* aws ssm parameter lookup - fail in case parameter doesn't exist
* aws ssm parameter lookup test case - failing case for nice return from path lookup
* aws ssm parameter lookup - convert incoming taglist to a key-value dictionary
* aws ssm parameter lookup - pep8 / style clean up
* aws_ssm lookup plugin rewrite for more standard interface
* aws_ssm module and lookup - introduce integration test and fix:
* aws_ssm module and lookup - error case integraton test and many PEP8 and other cleanups
* aws ssm parameter lookup - Various fixes in response to review + recursive fix & test
* aws ssm parameter lookup - more in response to review - shertel/abadger
* aws ssm parameter lookup unit test - move to mocker according to abadger
* aws ssm parameter lookup - integrate with new documentation fragment
* aws ssm parameter lookup - accept either aws_profile or boto_profile
* aws ssm parameter lookup - eliminate lookup document fragment until env vars are fixed later
This was causing an exception in the TaskQueueManager when a third
party handler plugin was processed that inherited or explicitly
called the callback method from super because the method signature
was incorrect in callback/__init__ and it processed the arguments as
incorrect data types as a side effect.
Signed-off-by: Adam Miller <admiller@redhat.com>
* Fix over-byte
* Update ios tests to call `provider`
To continue to support testing `connection: local`
* Fix command dict handling in ios_user
* Clean up unit tests, too
* Handle multiple sub prompts for network_cli connection
Fixes#35349
* Check if the same prompt is repeated in consecutive window
if it is repeated it indicates there is problem with answer
provided
* In that case report error to user
* Fix CI failure
* Fixes#35349
* Add prompt_retry count to control max number of times
to expect the same prompt before it error's out
* Make required changes in ios and eos terminal plugin to handle
wrong enable password correctly and return proper error
message to user.
* Check if the same prompt is repeated in consecutive window
if it is repeated it indicates there is the problem with an answer
provided
* In that case report error to user
Fixes#33993
In current implementation while trying to fetch the
current cli prompt terminal plugin executes `\r` on remote host.
This causes if command execution results in a prompt for
example executing `enable` command results in `Password:` prompt
there by executing `\r` on password prompt which is not intended.
To fix it while fetching current prompt from terminal plugin use `get_prompt()`
api in network_cli that returns the latest matched prompt instead of
unnecessary executing `\r` on remote host.
The rpc method should be exposed to the modules over the connection rpc
mechanism. This fix will expose the rpc method to call generalized rpc
endpoints on remote devices
* Fix edit_config multiline commands on IOS
The current code for multiline commands in IOS is broken: If you
pass a dict containing a command, prompt and answer it is seen
later as unicode string, but if you do a json.loads it fails
as the keys/values are enclosed in single quotes but JSON requires
double quotes.
Fixes#23539
* Fix pep8
* Use ast literal_eval
It's safe to use, as it is just for types and wont execute arbitrary code.
* refactors nios api shared code to handle provider better
This change refactors the shared code to be easily shared between
modules, plugins and dynamic inventory scripts. All parts now implement
the provider arguments uniformly.
This also provides a centralized fix to suppress urllib3 warnings coming
from the requests library implemented by infoblox_client
* fix up pep8 errors
* fix missing var name
* Add transfer checksum verification in copy module, to ensure that the file was transferred to the remote successfully. Fixes#35029
* Guard on no checksum
* Add version_added
* 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
* adds support for using connection=netconf
This change updates the module to provide support for using
connection=netconf instead of connection=local. If connection=netconf
is used, then the various connection arguments will be silently ignored.
* adds netconf plugin default
This adds a default implementation for netconf plugins if the network_os
is not specified. The default plugin will implement only the standard
netconf rpcs
* fix up pep8 issues
* Adding enos_config.py action plugin to create backup properly
* Changing BSD to GPL
* Changing License file to GPL
* Updating enos_config.py
* Updated copy right year to 2017
* Look for password prompts at the beginning of the line only
* Adds secondary checks for the supplied sub-prompts
Handles the case for enable prompts where enable_pass may be incorrect
and the deivce is still prompting for the password
* adds two new infoblox lookup plugins
* nios - lookup plugin to return nios objects to the playbook
* nios_next_ip - lookup plugin to return the next avaiable ip address
* adds some additional examples to nios lookup
* fix up pep8 failures
* updates in response to review comments
* Imported lookup plugin from Role
* Plugin cleanup, including:
* Use existing Python YAML parsing
* Remove environment variables as connection options
* Added initial debugging information
* Reworked the lookup plugin using the Python Request library. As it's available through Ansible, it makes communication with Conjur much more straight forward.
* Removed un-used libraries
* Fixed linting issues
* Standardized output on `format` and insure it works for 2.6, 2.7, and 3.x.
* Use quote_plus from the six library for improved python 2/3 behavior.
* Refactored identity & configuration to prefer user's file. This also includes a refactor to remove an un-needed dictionary merge method.
* Removed `requests` in favor of `ansible.module_utils.urls`.
* Refactored netrc loading to warn if host is not present.
* Tests and a refactor to support easier testing.
* Added reference to website
* Fixed two linting errors
* Fixed an extra line found by linting
* Updated file write to use binary to insure config files are written correctly
* Resolved linting issues
* Refactored config & identity loading to take advantage of plugin options
* Cleanup a bunch of small items caught by linting
* Removed extra line caught by linting
* Swapped in pytest and added some tests with mocked network responses
* Pushing to see if this approach works better...
* Refactored be open_url mocking based on feedback
* Fixed a couple linting issues & refactored mocking into each method to attempt to resolve a failing test
* Use a generic MagicMock for python 2.6
* Fixes doc typo
require -> required
* Use `type: path` in identity_file and config_file
Also removes `expanduser` calls below (which will now be called automatically on
paths.)
* Defines maintainers for conjur_variable plugin
* BOTMETA.yml:
** defines $team_cyberark_conjur as maintainers of Conjur Variable plugin
** adds myself and @jvanderhoof to that team
* Adds URLs to relevant documentation for Conjur Variable lookup plugin
* Clarifies "the server," "the machine" -> "controlling host"
The machine identity used is that of the Ansible controlling host, not any
server being provisioned or instructed. This documentation change aims to make
that relationship clear.
* Adds response code to exception message on authentication failure
* Enhances exception messages to specify the controlling host
These error messages are less likely to confuse a user as to which machine is
associated with the files, identities, and configurations being described.
* Adds ANSIBLE_METADATA for Conjur variable lookup plugin
The code that depends on this is all in the action plugins so we should
leave it there until we either move that action plugin code over
(fixup_perms2) or we give action plugins the ability to register new
config.
* Inventory caching
* Add inventory caching for virtualbox
* Don't populate cache for virtualbox with stdout, use a dict of inventory instead
* Fix error creating the cache dir if it doesn't exist
* Keep cache default False and set to True in VariableManager __init__
* Check all groups before determining if a host is ungrouped.
aws_ec2 is an inventory plugin intended to replace uses of the `contrib/inventory/ec2.py`
inventory script.
For advanced grouping, please use the `constructed` inventory plugin.
People expect to be able to upload files to s3 using standard
locations for files.
Providing an action plugin that effectively rewrites the `src`
key to the result of finding such a file is a great help.
Tests added, and IAM permissions corrected
* 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
Change cb1b705218 introduced the newline
parameter on network_cli plugin, but that was never introduced on cliconf.
This causes ios_user to break, since the newline value is never plumbed thru
to network_cli
* 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.
* Mellanox OS name change: MLNXOS changed to ONYX
Signed-off-by: Samer Deeb <samerd@mellanox.com>
* Fix alphabetical order of modules metadata
Signed-off-by: Samer Deeb <samerd@mellanox.com>
* WIP Refactor EOS code to use cliconf
* Fix connection.get where sendonly is True
* Fix pylint issue
* Remove return from send_config and various exec_commands
Also, removed a few try/except, which are anyways handled in
Connection.
* converted win_updates to an action plugin for automatic reboots
* do not set final result when running under async
* Updated documentation around win_updates with async and become
* 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
* Count ignored failed tasks and successful tasks with expected failure as failed tests.
* Make junit callback plugins changes backward compatible: add new task marker TOGGLE RESULT and new option fail_on_ignore
* update DOCUMENTATION for network_cli and netconf
This updates the DOCUMENTATION string for both the netconf and
network_cli connection plugin.
* add additional options to the connection documentation
* update documentation based on review and feedback
* adds persistent connection options to documentation string
Shell is implemented via the command module. There was a special case
in mod_args to do that. Make shell into an action plugin to handle that
instead.
Also move the special case for the command nanny into a command module
action plugin. This is more appropriate as we then do not have to send
a parameter that is only for the command module to every single module.
* Update task definitions for network_cli
* Add connection to debug messages
* Specify connection for prepare task
* pc won't be around for connection=network_cli
* Assorted Python 3 fixes
* Give default port if ansible_ssh_port missing
* delegate -> connection
* Extend error regex
* Add eos and fix tests to run multiple connections
* Update tests to report connection
* Add missing START messages
* Fix unspecified connection
* Python 3 updates
Exceptions don't have `.message` in Python 3
* Override `become` when using `connection=local`
* Slight restructuring to make eapi easier later on
* Move eapi toggle to prepare_eos
* Pull out connection on eapi tasks
* Add a lookup plugin for AWS account attributes
* PEP8
* Use config system instead of hard coding logic for environment variables
* simplify logic
* Return a flattened dict to make using easier
* Reformat return example
* winrm: attempting to get kerb auth to work on MacOS
* moved to use pexpect if possible as it is simpler
* Made the pexpect event more lenient around different localisations
* Fixes a prompt doubling issue that impacted missing ios enable password prompts.
Due to get_prompt sending a '\n' the prompts became doubled and out-of-sync with what
was expected. This caused the enable command prompts to be missed.
Also added verification that on_become succeeded to reach enable prompt.
* Moved prompt doubling comment per shippable
* 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
Provide Vault App role method to the lookup.
https://www.vaultproject.io/docs/auth/approle.html
Usage :
`{{ lookup('hashi_vault', 'secret=secret/hello:value auth_method=approle role_id=myroleid secret_id=mysecretid url=http://myvault:8200')}}`
You can skip `role_id` and `secret_id` if you set `VAULT_ROLE_ID` and `VAULT_SECRET_ID` environment variables.
* Fix vyos on network_cli on python3
bytes do not have format() in Python3
* Push connection to tasks, with bonus connection=local test
* Run tests without explicit connection set
* Add/update START messages where appropriate
* Change cast_list_to_dict to more generic rekey_on_member
cast_list_to_dict was taking an arbitrary data format in and returning
an arbitrary data format out. Rework this to be a more generic function
which creates a dict of dicts based on a member of the dict.
Remove cast_dict_to_list since rekey_on_member handles the use cases we
know about and cast_dict_to_list suffers from the same problems as
cast_list_to_dict. If this is still needed we could think about filters
we could add to do this in a short jinja2 pipeline.
* Fix bare excepts (bare excepts even catch sys.exit())
Previously if a child's group vars section was defined before the
child group itself, an edge case would be hit where the state of the
pending declaration would process as var and therefore drop the
child pending declaration context. This would result in the group
vars defined for the parent group being out of scope for the child
group.
Example:
[web:children]
appnodes
proxies
[web:vars]
deployment_type=prod
[appnodes:vars]
foo_var=true
[appnodes]
appnodes[1:3].example.com
[proxies:vars]
bar_var=true
[proxies]
proxies[1:3].example.com
Previously the deployment_type variable would be out of scope for
both the appnodes and proxies groups. This patch fixes that.
Signed-off-by: Adam Miller <maxamillion@fedoraproject.org>
* Fix _clean_result for debug callback with 1 char var names
The check in _clean_results was removing any keys that happened
to be one of the chars in ('invocation') instead of the string
'invocation'. This was meant to be a tuple but there was no comma
so the for iterated the string instead of the tuple.
Introduced in 9dba580204fcc7d95d4ca2ea2ae4ce23fe28c2ed
Update unit test to catch this.
Fixes#33723
* Use .pop() to remove invocation from results dict
In base callback _clean_results, simplify the way the
'invocation' item is removed.
Add some more unit tests.
* fixes guess_os for netconf connections
This change fixes invalid calls to play_context when the network_os is
not set and the connection attempts to guess the network_os. The method
will now check the correct values for ssh key file and allow agent
instead of returning errors.
* fix up pep8 issues
* - Netconf plugin addition for iosxr
- Utilities refactoring to support netconf and cliconf
- iosx_banner refactoring for netconf and cliconf
- Integration testcases changes to accomodate above changes
* Fix sanity failures, shippable errors and review comments
* fix pep8 issue
* changes run_command method to send specific command args
* - Review comment fixes
- iosxr_command changes to remove ComplexDict based command_spec
* - Move namespaces removal method from utils to netconf plugin
* Minor refactoring in utils and change in deprecation message
* rewrite build_xml logic and import changes for new utils dir structure
* - Review comment changes and minor changes to documentation
* * refactor common code and docs updates
* Adds a tmsh command fallback
The bigip terminal plugin assumes that the remote shell is tmsh.
This is not always true. The remote shell may be bash sometimes.
This adds a different shell command in the case that bash is the
remote shell
* Fixing upstream errors
* Refactor common network shared and platform specific code into package (part-1)
As per proposal #76 refactor common network shared and platform specific
code into sub-package.
https://github.com/ansible/proposals/issues/76
* ansible.module_utils.network.common - command shared functions
* ansible.module_utils.network.{{ platform }} - where platform is platform specific shared functions
* Fix review comments
* Fix review comments
* In order to maintain a single api call across
netconf and cli transport from module code change
signature of commit api to accept genric args.
* Fix doc string
Fixes#33406Fixes#33405
* Fix typo in network_cli for sendonly
* Send `abort` to remote device in case configuration fails
* Fix indentation issue in eos_static_route integration test
* Revert network_cli change
* Refactor VyOS to use cliconf
* Use show configuration commands on get_config
* Remove debug statement
* Construct command/answer/prompt if needed and fix commit comments
* Convert command/prompt/answer to bytes
* Warn on tests used as filters
* Update docs, add aliases for tests that fit more gramatically with test syntax
* Fix rst formatting
* Add successful filter, alias of success
* Remove renamed_deprecation, it was overkill
* Make directory alias for is_dir
* Update tests to use proper jinja test syntax
* Update additional documentation, living outside of YAML files, to reflect proper jinja test syntax
* Add conversion script, porting guide updates, and changelog updates
* Update newly added uses of tests as filters
* No underscore variable
* Convert recent tests as filter changes to win_stat
* Fix some changes related to rebasing a few integration tests
* Make tests_as_filters_warning explicitly accept the name of the test, instead of inferring the name
* Add test for tests_as_filters_warning
* Update tests as filters in newly added/modified tests
* Address recent changes to several integration tests
* Address recent changes in cs_vpc
* add type manipulation
* update for py3
* add tests for list to dict
* Add dict to list tests
* Update tests to search for regex in response
* pep8 clean up
* update Exception for py3
* update test to be py2/3 compat
* update for py26 compat
* potential fix for py3 and py26
* potential fix for py3 and py26 take2
* add new line to kick off shippable
* remove cache file created
* fix filter name
* add space for shipable
* Fix junos integration test fixes as per connection refactor (#33050)
Refactor netconf connection plugin to work with netconf plugin
* Fix junos integration test fixes as per connection refactor (#33050)
Refactor netconf connection plugin to work with netconf plugin
Fix CI failure
Fix unit test failure
Fix review comments
* 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
* Allow the user to circumvent adding -tt on ssh commands to help aid in
debugging ssh related problems.
* Move config to the plugin
* Set version_added
* Change yaml section to "connection"
* Fix ssh unit tests
This change will now track any created persistent connection and shut it
down at the end of the play run. This change also includes an update to
properly honor the reset_connection meta handler.
* Add new filter to parse xml output for network use cases
Fixes#31026
* Add parse_xml filter
* Add documentation for parse_xml filter
* Edited for clarity.
* Fix review comment and add unit tests
* Fix unit test CI failure
* Fix CI issues
* Fix unit test failures
* Fix review comments
* More copy edits.
- authorize was not working as expected
- use end command to get to exec context instead of exit
- error due to mishandling of global variable _DEVICE_CONFIG
* win_reboot: change to sample system uptime instead of checking port status
* added connection timeout back in as now we can manually set it per connection.
* some pep8 fixes
* fix up error message on timeout in case an exception wasn't fired
* Changed doc to English (US) and simplified uptime check
* moved conn timeout over to new config connection options
* 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
Noticed that there was just one old and one new place in the default
callback that did not use Display.display() to show data. Changed those
two places so that everything is uniform.
* Update connection play_context when socket exists
* Don't fail on connections other than network_cli
* Fix enable prompt detection on ios & eos
* Check against "Module not found" error code, defined in modules/jsonrpc.py
* Add prompt check in action plugin for network platform
In case of ignore_errors for a wrong configuration
the prompt is left in configuration mode and moved to
next task, if the next taks requires prompt to be
in operational state it results in failure.
Hence add a check to ensure right prompt at start of
each task run.
* Add prompt check in action plugin for network platform
* In case of ignore_errors for a wrong configuration
the prompt is left in configuration mode and moved to
next task, if the next taks requires prompt to be
in operational state it results in failure.
* Hence add a check to ensure right prompt at start of
each task run.
* Fix CI issue
* Fix CI issues
Fix review comment
Change iosxr exit command to abort as per review comment
* Do not run script in check mode
Fixes#30676
* Reformat script integration test
* Add integration tests for check mode of script module
* Fix name on test
* Cleanup temp file
* win_script integration test syntaxt changes
* Add check mode tests for win_script
* Use proper variable in test
* Fail if source file does not exist
* Verify script is accessible and don't copy in check mode
Use shlex to properly split shell arguments, though a path with spaces in it still needs to be quoted in the playbook.
Add note to docs describing such.
Improve error message if file is not found indicating there may be a space in the path.
* Properly encode path now that path is split using shlex
* Allow for spaces in both path and script name
* Add unicode character test to Linux script tests
* Add Linux test for space in path to script
* fixes asa action plugin for connection=local
This change fixes asa modules when using connection=local to load the
provider values.
* fix up pep8 issues
* Remove uses of assert in production code
* Fix assertion
* Add code smell test for assertions, currently limited to lib/ansible
* Fix assertion
* Add docs for no-assert
* Remove new assert from enos
* Fix assert in module_utils.connection
* update sros local action plugin to support network_cli
This updates the sros local action plugin to only start the connection
if connection=local is specified. This is to support network_cli
connection plugin
* fix up pep8 issues
* Automatically loads and executes an inventory plugin specified by a standard YAML inventory config file containing a `plugin` key at its root.
* Moved inventory PluginLoader to a shared global instance.
* fixes to events/callbacks
- made note of 'not called' methods for future fixes
- removed uncalled v2_runner_on_file_diff because dupe of v2_on_file_diff, which is called
- removed v2_runner_on_no_hosts due to existing pb level ones, which are called
- removed v2_on_setup, it is just a task, triggers normal task events
- v2_on_notify is now called when a handler is notified
- TODO: async, cleanup? and import events
these currently occur in code that has no access to sending events
* corrected display
* 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
* Squashing all commits to one as suggested by John
* Adding Unit test method for the module enos_facts.py
* Pep8 and Ylint issues addressed
* Trying again to remove blank line. Some scripts are required for this.
* Bug Fixing for interfaces
* Editing for over indenting issue
* E203 whitespace before ','
* Update enos.py
Added warnings argument as to check_args method
* Update enos_facts.py
Added warnings to check_args method
* Add yaml output plugin
Using YAML instead of JSON for detailed output can (IMO) improve
readability; especially for tasks with either lots of output, or
multi-line output.
* Fix yaml output plugin documentation
* Added copyright header to yaml.py
* Moved mention of yaml output plugin to `New Modules` section
* Use AnsibleDumper in yaml output plugin
Also moved the `represent_scalar` setup into the init method. As a
global statement, it was causing exceptions trying to get an `id`
field that does not exist.
* Remove trailing whitespace
* Remove useless json.loads/dumps from yaml output
* removed/blobified unused PInvoke stuff
* added try/finally around impersonation to ensure RevertToSelf is called in all cases
* added a few explanatory comments
* win_become: make it easier to become with an admin token
* Fixed up pep8 whitespace
* fix for Server 2008
* Added support for async and become on newer hosts and fix warnings
* Enable ECHO in prompt module
Fixes#14160
* Add option for controlling echo behavior with pause module
* Improve option logic
Allow all options to be used in varying combinations, rather than being mutually exclusive.
Always capture output and return it, even when a time limit is set.
* Add version_added to docs
* Improve behavior of echo output
Set a few more flags to allow interactive deletion and hide control characters.
Do not capture or echo input when a time is set. Tried to get this working nicely, but ran into too many issues/oddities to keep it. Maybe in the future if there is demand for capturing/echoing input when a time is set I'll take another pass at it.
* show specific undefined var errors on -v in debug
allows users to get more specific information about undefined errors
as they might be looking at a complex data structure and need to find
the specific leaf that has the issue
* now tests works at all verbosity levels
* updated to unicode objects
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
avoid making gathered facts high precedence, only set_fact is supposed to be.
vars set via set_fact with cacheable are higher precedence than plain facts.
Previously (after 6fbd0a8bb5) regular facts would end up with a
higher precedence than host or play vars, but they should not be. Facts were getting added to 'non_persistent_facts' (equivalent to 'register' vars) which is higher precedence than facts should be.
added 'cacheable set_facts' to precedence docs
'ansible_facts_cacheable' -> '_ansible_facts_cacheable' (made 'private')
* Fix wrong prompt issue for network moodules
Fixes#31161Fixes#32416
* Store the device prompt in case of error
from remote device
* Check for prompt value in ios action plugin
* Add integration test
* Enable ECHO in prompt module
Fixes#14160
* Set flags to make it possible to edit echoed input as well as hide control charcters
Only do this if a time limit is not set.
* Consolidate settings
The eos terminal plugin did not correctly catch the error message
returned with trying to configure more than one ospf instance. This
change updates the terminal plugin to catch that scenario
* 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
* better cleanup on task results display
callbacks get 'clean' copy of result objects
moved cleanup into result object itself
removed now redundant callback cleanup
moved no_log tests
* moved import as per feedback
* Adding a cli transport option for the bigip_command module.
* Fixing keyerror when using other f5 modules. Adding version_added for new option in bigip_command.
* Removing local connection check because the F5 tasks can be delegated to any host that has the libraries for REST.
* Using the network_common load_provider.
* Adding unit test to cover cli transport and updating previous unit test to ensure cli was not called.
* Create action file enos_facts.py
* Update and rename enos_facts.py to enos.py
* Taking chances on Dealing with Unstable issues
* Removing blank space/ white line
* fixed module generation
added missing lookup page
point to plugins when plugins
made modules singular
add display for verbose an debug messages
nicer templating, changed generation order for ref
corrected links
moved most of lookup docs to plugin section
* Copy edits
* Fixed typos
* Clarified wording
* win_become: Added support to become a service user
* fixes for linting
* changes to get local and network service working
* fixed linting issues again
* pleasing pepe
- new module: ssm_parameter_store
- new lookup: ssm
* lookup module ssm - adjust error message
* Pacify pylint erroring on botocore not found
* adjust to version 2.5
this was abandoned early on the manger side but seems like we left behind on plugin side.
more flexible extensions with yaml plugin
validate data correctly for yaml/constructed
fixed issue with only adding one child to keyed, the group only got the host that forced it's creation
fixes#31382fixes#31365
* win_become: move error handling to Ansible outside of shell
* trimmed the output so double newlines don't get set
* added test for non-zero exit code
* missed issue URL on test
* changed exit to SetShouldExit
* updated docs
- for devs:
- added inventory/vars section
- made some updates to general section and other plugin types
- for users:
- added 'user' plugin section to start describing the plugins
- docs on types, what they are and how to use
- removed ref to deleted AUTHORS file
- corrected several typos/headers
- added descriptions to config.rst template
- ignore generated files for cli/plugins and config
- remove new generated files on `make clean`
- moved details from devguid and intro doc to plugin specific pages
- pretied up lookup notes
- changed precedence ref to not conflict config
- removed duplicate config data, as config is autogenerated and up to date
- put new plugins under playbooks
- added `pass` cause rst/python dislikes fractions
- removed dupe in .gitignore, alpha sorted to avoid moar dupes
- added try cause rst/python freaks out
* generate plugins into their own dir
only do plugins that support docs
use toctree from main plugins page
* Fix nxos provider transport warning issue
* Add default value of transport arg in provider spec
* Remove default value if transport arg in top level spec
This ensure deprecation warning is seen only in case transport
is given as a top level arg in task
* Refactor nxos modules to reference transport value from provider
spec
* Fix unit test
* Remove transport arg assignment in nxos action plugin
* As assigning transport value is handled in provider spec
top level task arg assignment is no longer required
This PR includes:
- Support for loop-tasks with proper subject/error content
- Improved output (and proper indentation)
- Complex data structures are now pretty printed
- Better selection of mail subject
* Restore correct coloring to selective callback
This fixes the bug raised in #30506
* Fix format issues for Python 2.6 & indent
Removed the zero length fields to support format under Python 2.6
Fixed E128 continuation line under-indented for visual indent issue
* finalize lookup documentation
* minor fixes to ansible-doc
- actually show which file caused error on when listing plugins
- removed redundant display of type and name
* smart quote fixes from toshio
* fixed issue with default callback inheritance
- callbacks need to document same options as callbacks they inherit from to get them configured
- since default is also used by many 3rd party callbacks for inheritance, making the code 'tolerate' the missing docs
and fallback to using the direct constant to configure it's options.
The dellos action plugins should add the remote address of the switch
provider to the play context. This was fixed in issue #23589 in an
almost identical manner for the eos, ios, iosxr, and vyos action
plugins.
Fixes: #30350
* nxos_file_copy bug fix
Signed-off-by: Trishna Guha <trishnaguha17@gmail.com>
* provider gets set to None in module level when transport is cli
Signed-off-by: Trishna Guha <trishnaguha17@gmail.com>
Fixes#29060
Allow delegate_to if transport is either nxapi or eapi.
Persistent connection uses `cli` transport and create
a local socket on control node. Hence delegate_to is not allowed
for `cli` transport.
However as `nxapi` and `eapi` transport does not use persistent connection
it is possible to use delegate_to in this case.
* Fix unwanted deprecation message in network module argspec
Fixes#25663Fixes#24537
* segregate provider spec and top level arg spec
* add deprecation key in top level arg spec
* remove action plugin code to load provider and add
that logic at a common place in network_common.py file
* Fix CI issue
* Minor change
* Using docstrings conflicts with the standard use of docstrings
* PYTHON_OPTIMIZE=2 will omit docstrings. Using docstrings makes future
changes to the plugin and module code subject to the requirement that we
ensure it won't be run with optimization.
* module and vault fixes
- fix module_path cli option and usage, which fixes#29653
- move --output to be in subset of vault cli, no need for all vault enabled cli to use it
- added debug to loader to see directories added
* only complain about ini deprecation if value is set
* set plugin config for stdout and other types
* updated plugin docs, moved several plugins to new config
* finished ssh docs
* fixed some issues seen in plugins while modifying docs
* placeholder for 'required'
* callbacks must use _plugin_options as _options already in use
from __future__ unicode_literals leads to developer confusion as
developers no longer can tell whether a bare literal string is a byte
string or a unicode string. Explicit marking as u"" or b"" is the way
to solve the same problem in the Ansbile codebase.
- better variable precedence management
- universal plugin option handling
- also updated comments for future directions
- leverage fragments for plugins
- removed fact namespacing
- added 'firendly name' field
- updated missing descriptions
- removed some unused yaml entries, updated others to reflect possible future
- documented more plugins
- allow reading docs using alias
- short licenses
- corrected args for 'all plugins'
- fixed -a option for ansible-doc
- updated vars plugins to allow docs
- fixed 'gathering'
- only set options IF connection
- added path list and renamed pathspec mostly the diff is , vs : as separator
- readded removed config entries that were deprecated but had no message ... and deprecated again
- now deprecated entries give warning when set
* cleaner get for file based caches
* now db based facts behave like file ones
we now keep local in mem cache to avoid race conditions on expiration during ansible runs
* Fix encoding errors on grp.gr_name, which can contain non-ascii character at LDAP/AD domain workstations
* fix: utils.to_text() is now used instead of py3-incompatible unicode() method
So 5 years after the original version, it's time to give this callback
plugin some needed rework.
This PR includes:
- Reuse same logic for failed, unreachable or Async errors
- Add playbook and task information at the top (easier to find context)
- Improve the sections (ie. indented content)
This is part of the effort to make windows parameters conform to
other modules. Usually parameters don't include the unit in the
parameter name.
See also #20160
This PR includes:
- Check-mode support, the module will now report as if it was rebooted,
without making the actual reboot.
- Async-support, this potentially fixes#23835
- The module now also returns the elapsed time, like the wait_for and wait_for_connection modules.
This fixes#18108
- Add post_reboot_delay_sec parameter
