* Fix incorrect parsing of header containing equal character
* Add changelog fragment
* Update changelogs/fragments/7303-mail-incorrect-header-parsing.yml
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* Fixed gitlab_*_variable when using purge, see issue #7250
* Update changelogs/fragments/7251-gitlab-variables-deleteing-all-variables.yml
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* gio_mime: new module
* multiple changes
- fixed gio_mime module utils handler detection
- add message to test helper assertion
- add unit test
* gio mime output uses unicode smart quotes
* test helper: do not use skip/xfail reason messages - break in old pytest
* Update plugins/modules/gio_mime.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* change exist variables to meet standards
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add keycloak_realm_key module
* keycloak_realm_key: make "ansible-test sanity" happy
Signed-off-by: Samuli Seppänen <samuli.seppanen@puppeteers.net>
* keycloak_realm_key: support check_mode
* keycloak_realm_key: add integration tests
* keycloak_realm_key: remove FIXME comment
* keycloak_realm_key: fix EOL in integration test variables
* keycloak_realm_key: remove unused import
* keycloak_realm_key: remove integration test realm at the end of test suite
* keycloak_realm_key: add version_added metadata
* keycloak_realm_key: add documentation for end_state
* keycloak_realm_key: support the "certificate" parameter
As with "private_key" changing the certificate after creation is not possible
because we can't compare the current value to the desired value.
* keycloak_realm_key: document default for certificate parameter
Signed-off-by: Samuli Seppänen <samuli.seppanen@puppeteers.net>
* keycloak_realm_key: implement diff mode
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* keycloak_realm_key: remove note about literal linefeeds
* keycloak_realm_key: remove defaults from priority and certificate
* keycloak_realm_key: mark diff and check modes as partially supported
* keycloak_realm_key: implement "force" parameter
This ensures that the desired state is always enforced on keys that should be,
and are, present.
* keycloak_realm_key: fix yaml parsing error in documentation
* keycloak_realm_key: document why check_mode support is partial
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* keycloak_realm_key: documentation and metadata fixes
* keycloak_realm_key: small documentation fix
* keycloak_realm_key: change version_added to 7.5.0
* Update plugins/modules/keycloak_realm_key.py
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Signed-off-by: Samuli Seppänen <samuli.seppanen@puppeteers.net>
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update as per PR comments
* Move common code to module_utils
* Break up long import line
* Fix pipeline errors
* Inital version of check_mode support
* Fix updating a role, add tests
* Fix line spacing
* Fix line indentation
* Add consul-role tests
* Fixes for role update
* Apply suggestions from code review
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update as per MR comments
* Update as per MR comments
* Fix documentation issues
* Add types for sub-options
* Allow setting of policy, service and node id fields by specifying a value, or leaving them unchanged by omitting them
* Fix typo in test
* Apply suggestions from code review
Co-authored-by: Felix Fontein <felix@fontein.de>
* Reset and force push to get rid of merge
* Corrected unit tests
* Apply suggestions from code review
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add suboptions documentation for node and service identities
* Fix PEP errors from pipeline
* Fix pipeline errors.
* Fix more pipeline errors
* Apply suggestions from code review
Co-authored-by: Felix Fontein <felix@fontein.de>
* Fix line that is too long
* Not specifying a value for description during update now leaves existing value unchanged
* Fixes for pipeline errors
* Add test cases to verify handling description works
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* Edit givenname and sn description.
* Update plugins/modules/ipa_user.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update plugins/modules/ipa_user.py
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add `ignore_spaces` option to `ini_file` to ignore spacing changes
Add a new `ignore_spaces` option to the `ini_file` module which, if
true, prevents the module from changing a line in a file if the only
thing that would change by doing so is whitespace before or after the
`=`.
Also add test cases for this new functionality. There were previously
no tests for `ini_file` at all, and this doesn't attempt to fix that,
but it does add tests to ensure that the new behavior implemented here
as well as the old behavior in the affected code are correct.
Fixes#7202.
* Add changelog fragment
* pep8 / pylint
* remove unused import
* fix typo in comment in integration test file
* Add symlink tests to main.yml
It appears that #6546 added symlink tests but neglected to add them to
main.yml so they weren't being executed.
* ini_file symlink tests; create output files in correct location
* Add integration tests for ini_file ignore_spaces
* PR feedback
* nsupdate: fix 'index out of range' error when no TTL answer is given
Fix a possible `list index out of range` when no answer is returned in the `ttl_changed` method
by applying the existing workaround for NS records to all record types.
Resolves#836
* fixup! nsupdate: fix 'index out of range' error when no TTL answer is given
* Add detach option so the task doesn't wait until the Jenkins job is finished
* Add new time_between_checks to be able to configure the sleep time between requests
* New jenkins_build_info to get information about a specific build
* Add version_added to the new module
* Add changelog fragment for jenkins_build changes
* Fix tests that required the python-jenkins module
* Remove tests that failed
Doesn't really make sense to test that with a mock
* Fix pep8 error
* Update maintainers for the jenkins_build and jenkins_build_info modules
* Improve format and add link to PR
* Move version_added documentation to the right file
* Fix incorrect examples
* Improve text format
* Fix incorrect code style
* Fix incorrect YAML for documentation
* Add version_added documentation to the new options
* (feat) New module pnpm added
A new module for pnpm is added. Necessary entries were written to
BOTMETA.yml.
* (feat) Basic tests added
* (feat) reduced nesting of ifs
* (fix) trying to fix up CI pipelines
* (fix) incorrect indentation of alias fixed
* (feat) fixed further indentations
* Apply suggestions from code review
Co-authored-by: Felix Fontein <felix@fontein.de>
* (fix) various linting and CI errors fixed/ignored
* (feat) reduced restriction, new install method
Some restrictions on OS are reduced. New installation method, similar to
the official installation method, is provided.
* (fix) ignoring CentOS 6 for CI.
* retrigger checks
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* community.general.make:
allows parameters without value
closes#7178
* add changelog fragment for community.general.make
* correction: v != none -> v is not None
* update fragment changelog as per developer request
* add an example
* document the modification
* update example with comments as per maintainer request
* add support for percentage of origin size for creating snapshot volumes
* add changelog fragment
* add pull request link
Co-authored-by: Felix Fontein <felix@fontein.de>
* fix what's not idempotent
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* sorcery: update only specified grimoires
* Update plugins/modules/sorcery.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add a flag to omit listing new repositories before add/remove
* No need to append an empty string
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* [proxmox_vm_info] Return empty list when requested VM doesn't exist
* Update documentation
* Add changelog fragment
* Address review comments
* Allow to filter by empty name
* Update plugins/modules/proxmox_vm_info.py
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add grimoire management to sorcery module
* Add changelog fragment
* Bump copyright year
* Separate update_cache and latest state
* Apply suggestions from code review
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add note on latest state and cache_update link
* Unblock execution of multiple stages
* Update plugins/modules/sorcery.py
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update Codex logic to match Sorcery
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* ejabberd_user: bug fixes + tests
* fix changed property
* add license to handler file
* adjustments to test
* add needs/target/setup_epel to aliases
* further adjustments to integration tests
* add target to integration tests
* add some skips to test
* skip centos as it has no ejabberd
* skip fedora as it has no ejabberd
* discard unused epel setup
* add changelog frag
* remove ejabberd before tests
* fix typo
* Only treat `---` as an info separator when it's preceded by newline
The code for splitting the output of `snap info` for multiple snaps
can't assume that `---` separates snaps any time it appears in the
output; it needs to treat that as the separator only when it's at the
start of a line. Otherwise it breaks if any snap uses `---` in its
description text, which, e.g., the `bw` snap does as of the date of
this commit.
Fixes#7045.
* Add changelog fragment
* Add a comment explaining why \n is necessary before ---
* Update changelogs/fragments/7046-snap-newline-before-separator.yml
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Jonathan Kamens <jik@jik5.kamens.us>
Co-authored-by: Alexei Znamensky <103110+russoz@users.noreply.github.com>
Co-authored-by: Felix Fontein <felix@fontein.de>
* Add example for ECS Fargate/EFS Jenkins authentication
Since ECS Fargate is serverless, one cannot access its jenkins_home other than from a machine (EC2 for example) that actually mounts and owns its EFS storage.
That way we provide user/group of a defatul local user which has the same uid/gid 1000 as the default jenkins user inside the container and also can authenticate at Jenkins URL.
I feel this is not as straightforward from the docs and someone might benefit from such an example being present
* Added an empty line
* Float value now in single quotes
* Use UID/GID instead user/group name
* enable 'force' restart of vm
* added changelog fragment
* Update changelogs/fragments/6914-proxmox_kvm-enable-force-restart.yml
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
* [proxmox_vm_info] Re-use cluster resources API to use module without requiring node param
* More consife if
* Fix use case when requesting all vms from specific node
* Add changelog fragment
* Fix key error for reverse zone (#6905)
* Add changelog fragment for #6905
* Fix changelog (6905)
Co-authored-by: Felix Fontein <felix@fontein.de>
---------
Co-authored-by: Felix Fontein <felix@fontein.de>
Simplify a bit (and possibly speed it up a little) the parsing of the
output of `subscription-manager repos --list`:
- simplify skipping the lines that are not interesting: check the first
character only, as it is enough to determine whether it contains
repository data or not
- check the start of each line manually, rather than with regexp: a
simple slice + lstrip() gives the same result
