* basic plugin loading working (with many hacks)
* task collections working
* play/block-level collection module/action working
* implement PEP302 loader
* implicit package support (no need for __init.py__ in collections)
* provides future options for secure loading of content that shouldn't execute inside controller (eg, actively ignore __init__.py on content/module paths)
* provide hook for synthetic collection setup (eg ansible.core pseudo-collection for specifying built-in plugins without legacy path, etc)
* synthetic package support
* ansible.core.plugins mapping works, others don't
* synthetic collections working for modules/actions
* fix direct-load legacy
* change base package name to ansible_collections
* note
* collection role loading
* expand paths from installed content root vars
* feature complete?
* rename ansible.core to ansible.builtin
* and various sanity fixes
* sanity tweaks
* unittest fixes
* less grabby error handler on has_plugin
* probably need to replace with a or harden callers
* fix win_ping test
* disable module test with explicit file extension; might be able to support in some scenarios, but can't see any other tests that verify that behavior...
* fix unicode conversion issues on py2
* attempt to keep things working-ish on py2.6
* python2.6 test fun round 2
* rename dirs/configs to "collections"
* add wrapper dir for content-adjacent
* fix pythoncheck to use localhost
* unicode tweaks, native/bytes string prefixing
* rename COLLECTION_PATHS to COLLECTIONS_PATHS
* switch to pathspec
* path handling cleanup
* change expensive `all` back to or chain
* unused import cleanup
* quotes tweak
* use wrapped iter/len in Jinja proxy
* var name expansion
* comment seemingly overcomplicated playbook_paths resolution
* drop unnecessary conditional nesting
* eliminate extraneous local
* zap superfluous validation function
* use slice for rolespec NS assembly
* misc naming/unicode fixes
* collection callback loader asks if valid FQ name instead of just '.'
* switch collection role resolution behavior to be internally `text` as much as possible
* misc fixmes
* to_native in exception constructor
* (slightly) detangle tuple accumulation mess in module_utils __init__ walker
* more misc fixmes
* tighten up action dispatch, add unqualified action test
* rename Collection mixin to CollectionSearch
* (attempt to) avoid potential confusion/conflict with builtin collections, etc
* stale fixmes
* tighten up pluginloader collections determination
* sanity test fixes
* ditch regex escape
* clarify comment
* update default collections paths config entry
* use PATH format instead of list
* skip integration tests on Python 2.6
ci_complete
* Add a force_replace_host flag to win_domain_membership
Satisfies https://github.com/ansible/ansible/issues/53539
* Rework backticks
* Bump version_added
* Check for existence of current hostname as well; use LDAPFilter during search
* Rename $force_replace_host to $allow_existing_computer_account
* Added docs, porting guide and minor nit in code
* Catch all request timeouts for winrm connection
The current implementation only catches 'ConnectTimeout' exceptions.
Instead we should catch 'Timout' which also catches ReadTimeout
exceptions.
Improves on: #51744
Co-Authored-By: westphahl <westphahl@gmail.com>
* Changelog for winrm error handling improvement
* Use locking for concurrent file access
This implements locking to be used for modules that are used for
concurrent file access, like lineinfile or known_hosts.
* Reinstate lock_timeout
This commit includes:
- New file locking infrastructure for modules
- Enable timeout tests
- Madifications to support concurrency with lineinfile
* Rebase, update changelog and tests
We need to specify ansible_python_interpreter to avoid running interpreter discovery and selecting the incorrect interpreter.
Remove the import of lock in known_hosts since it is not used.
* adding (optionally) image information to inventory var
* add boot image mapping to gcp_compute instance data for all disk
image data in the configured zones
Signed-off-by: Adam Miller <admiller@redhat.com>
* start fixing gcp inv plugin
* minor fixes
* added clog
* ajust comments
* link indv zone/project
* separate specific zone/project from other params
* restoring zones query per project
* also work when zones given
* fixed scopes, removed incorrect docs as not option
* Added support for types parameter
- Parameter is used to specify multiple network types
* Fix documentation
* Apply suggestions from code review
Co-Authored-By: kbreit <kevin.breit@kevinbreit.net>
* Reworked type parameter to be a list so types isn't needed
* Re-add tags documentation
* Fix documentation around compatibility
* Convert tags to list from string
* Add changelog fragment
When using before and after in combination, the opposite behavior was induced. This PR makes the the replacement happen between the specified patterns as intended.
* Added integration tests
* Add changelog, porting guide entry, and minor doc fixes
* Removed required_together, updated tests
Since required_together: privatekey_path -> friendly_name, is not always
required it has been removed.
Updated openssl_pkcs12 integration tests to be in line with other
openssl_* modules, and added a test for export with no privatekey_path.
* linter fixes
* Removed cryptography from tests
* Added changelog fragment
* Removed non-necessary select_crypto_backend
* fix AWS plugin credential precedence for environment variables
* Allow aliases in direct plugins options
Consolidate precedence fix just in the doc fragment using aliases for mismatched options
* Access options with the option name rather than alias
* fix indentation
* update unit tests
* Improve readability
* Adding networks_cli_compatible option.
* Move network tests into own test file.
* Extend tests (for networks_cli_compatible=no).
* Adding tests for networks_cli_compatible=yes.
* There seems to be no way to create a container without at least one network attached.
* Integrate networks / purge_networks with comparisons.
* Speed up tests.
* Removing double dot.
* Add changelog.
* Use comparisons value only if the networks option has been specified. purge_networks on the other hand also removes networks if it has not been specified.
* Add write helper.
* Adjust modules (except openssl_certificate).
* Adding tests for mode (with openssl_privatekey).
* Add openssl_certificate support.
* Never, ever remove the output file before actually trying to generate new content for it.
Removal is only allowed when state=absent, or when the object has been regenerated and the result needs to be written to that place.
* Add changelog.
* Extend test.
* New cryptography backend for openssl_certificate
load_* functions in module_utils/crypto.py now have a backend paramter
which when set to 'cryptography' will return cryptography objects so
they can be used for both pyopenssl and cryptography backends.
Added a select_message_digest function too returning a cryptography
digest hash from `cryptography.hazmat.primitives.hashes`
Added new classes for Cryptography backend
* Run test with various backends.
* Prefixing tests.
* Make sure we have the correct backend available.
* Linting (flake8).
* Moved cryptography import to separate try/except
* Make sure certificate is actually valid at some time in the past.
* Improve error handling.
* Trying to fix validation for cryptography backend.
* Fixed issue with keyUsage test in assertonly
* Fixed CI/Lint issues
* Fix private key problem for OwnCA.
* Cryptography backend doesn't support v2 certs.
* issue an expired cert with command when using cryptography backend
* Added warning when backend is auto and v2 cert is requested
* Bumped min cryptography version to 1.6
* Correctly check for failure when backend is cryptography and cert is v2
* Use self.backend where possible
* Use secp521r1 EC when testing on CentOS6
* Fixed pylint issue
* AcmeCertificate support for both backends
* Review fixes
* Fixed missing '(' when raising error
* Fixed date_fmt loop
* Updated docs and requirements with cryptography
* Add openssl_certificate to changelog.
Fixes#40060
* Fix coding style errors
* Use CONNECTION LIMIT (no underscore)
* From review done by amenonsen and bcoca - Set default at None, make the change detection less confusing
* Added EXAMPLE on how to apply a database specific connection limit
* Added some basic tests for conn_limit applied to a database
* Check that conn_limit has actually been set / updated to 200
* Add changelog fragment regarding postgresql_db conn_limit parameter
* cs_volume: add volumes extraction and upload features
* cs_volume: Update doc, remove deprecated code
* cs_volume: Add unit tests for extract and upload features
* Fixes flatpak module to work with flatpak >=1.2.0, fixes#51481
This keeps backwards-compatibility for flatpak versions before 1.2.0
* Fixes typeo
Co-Authored-By: oolongbrothers <oolongbrothers@zeibar.net>
* #50877:
* add support to postgresql_privs to use "FOR { ROLE | USER } target_role"
in "ALTER DEFAULT PRIVILEGES"
* fix sanity errors
* #50877: fix documentation and add a check for correct usage
of target_roles
* #50877: fix missing absent option for default privs with target_role
* #50877: add clear description, when target_roles can be used
* #50877: fix conflicts, formatting, and add a changelog fragment
* #50877: fix sanity error E335
* #50877: swap conditions and fix error to warning msg
* #50877: add tests for default privileges
* #50877: fix tests for default privileges
* #50877: fix tests for default privileges on centos 6
The output of pw.getpwnam() does not distinbuish between local and remote accounts. It will return a result if an account exists locally or in the directory. When local is set to True in the task parameters, look through the local password database explicitly.
* Ensure luseradd is present for tests
* Add docs and warnings about local mode
* Allow parent groups to be variables or literal, requires {{ }}
* Check strict before failing on templating errors
* Don't add a group if an invalid parent group was provided
* avoid using Postgres formatting function
* add tests for ALL FUNCTIONS IN SCHEMA
* documentation and changelog
* requested changes in tests
* fixed changelog
* Add source option.
* Split force parameter into force_source, force_absent and force_tag.
* Move all build-related options into a suboption called build.
* Add changelog.
In pyyaml versions before 5.1 the default_flow_style for yaml.dump
was None. Starting with 5.1 it is now False. This change explicitly
sets the value to None to maintain the original to_yaml behavior.
The change to pyyaml was made in the following commit:
507a464ce6
When user specifies the JUMP value to 'tee', gateway is required.
This fix adds new parameter 'gateway' to support this functionality.
Fixes: #53170
Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>
* Adding support for IP identifiers according to https://tools.ietf.org/html/draft-ietf-acme-ip-05.
* Add changelog.
* Make sure that the authorizations return value is unchanged for CSRs with DNS-only SANs.
* Remove unneeded import.
* type -> identifier_type
* Python 2.6 compatibility.
* Fix unit tests.
* Add IP address normalization.
* Extend tests.
* Move data into fixtures.
* Adjust BOTMETA.
function changed to do in place replacement, should be less expensive even with copy as it avoids 'sub copies', can compose with module_args_copy to create replacement for old behavior
attempt to fix#52910
* handle lists and subdicts correctly
* added missing exception case, which was not noticed since 'cleaning' was not working
* added comments to clarify exceptions
* create overridable sanitation function
* now used in aws, gce and azure plugins
* added new option to these plugins to work in conjunction with general toggle to make it easier
to emulate inventory script behavior.
* Don't raise AnsibleConnectionFailure if the ssh_process has already died. Fixes#53487
* Better support for file not found messages
* Add changelog fragment
* changelog
* combine provided variables and host vars inside of constructing groups to take into account composed variables
let composed variables "win"
* fix whitespace
* Allow user to control hash behavior
* Remove dependency to psycopg2 with dump/restore
'dump' and 'restore' state only need pg_dump and pg_restore. These tools
don't use psycopg2 so this change tries to avoid the use of it in these
cases.
The db_exists test was replaced with an error detection when piping to
compression program, using a FIFO file. This effectively reverts #39483,
that was a fix for #39412.
* Fix typo
* Add changelog fragment
* Add note for dump and restore not requiring psycopg2
* Fix YAML syntax
* Update lib/ansible/modules/database/postgresql/postgresql_db.py
Co-Authored-By: Glandos <bugs-github@antipoul.fr>
* Output warnings from docker daemon on container create and update.
* Accept warning for blkio_weight instead of idempotency.
* Value quoting.
* Avoid loop variable conflict.
* Add changelog.
* Make one test case faster.
* Add 'Docker warning: ' prefix.
* Add a generalized warning reporting function.
* Add rollback_config
* Add change log fragment
* Fix broken test
* Actually fix broken tests
* Add rollback_config example
* Default rollback_config as None
* Abort early if rollback_config does not exist
* Add mount options
* Remove mount readonly default
* Fix driver_config test
* Add documentation
* Add changelog fragment
* Properly indent tmpfs_ options
* Use correct service suffix for mount tests
* Check for None value on tmpfs usage check
* Document change of mounts.readonly return key
* Use correct change log type
* Really use correct change log type
* Revert changing mount.readonly to read_only
* Raise OpenSSLBadPassphraseError if passphrase is wrong.
* Improve handling of passphrase errors.
Current behavior for modules is: if passphrase is wrong (or wrongly specified), fail.
Current behavior for openssl_privatekey is: if passphrase is worng (or wrongly specified), regenerate.
* Add changelog.
* Add tests.
* Adjustments for some versions of PyOpenSSL.
* Update lib/ansible/modules/crypto/openssl_certificate.py
Improve text.
Co-Authored-By: felixfontein <felix@fontein.de>
Fixes a bug where parse_distribution_file_ClearLinux() was called on CoreOS (and probably many other distros) and it returned True since it successfully parses the distribution file. Since this file exists on many Linux distributions and they are a very similar format, add an additional check to make sure it is Clear Linux.
Change the order in which distribution files are processed so NA is last. This prevents a match on CoreOS hosts since they also have /etc/os-release and the called matching function for NA is very general and will match CoreOS.
* Add changelog
* Add unit tests
Only add tests for Clear Linux parsing since that was the cause of this issue.
Matt Davis