When loading variables in all ansible-freeipa roles, it is expected
that a file with these variables is present for each supported Linux
distribution, and then, based on the information about the distribution
provided by Ansible, the correct file is loaded.
Previously, only the facts `distribution` and dinstribution version
related facts were used, which required specific files, or links to
files for distributions in the same "family", which will probably have
the same variables set.
This change adds searching for files based on the `os_family` fact,
allowing distributions that follow the same family rules to be
supported, without any changes to the codebase. It is still possible
that a specific distribution configuration overrides the default
behavior, as `os_family` has lower priority than `distribution`.
For example, distributions on the `RedHat` family, like Oracle Linux,
Alma Linux, and Rocky Linux, work withoutadding new files, or links to
files, to fill the `vars`.
Fix issue #573. Fix issue #523.
When loading variables in all ansible-freeipa roles, it is expected
that a file with these variables is present for each supported Linux
distribution, and then, based on the information about the distribution
provided by Ansible, the correct file is loaded.
Previously, only the facts `distribution` and dinstribution version
related facts were used, which required specific files, or links to
files for distributions in the same "family", which will probably have
the same variables set.
This change adds searching for files based on the `os_family` fact,
allowing distributions that follow the same family rules to be
supported, without any changes to the codebase. It is still possible
that a specific distribution configuration overrides the default
behavior, as `os_family` has lower priority than `distribution`.
For example, distributions on the `RedHat` family, like Oracle Linux,
Alma Linux, and Rocky Linux, work withoutadding new files, or links to
files, to fill the `vars`.
Fix issue #573. Fix issue #523.
When loading variables in all ansible-freeipa roles, it is expected
that a file with these variables is present for each supported Linux
distribution, and then, based on the information about the distribution
provided by Ansible, the correct file is loaded.
Previously, only the facts `distribution` and dinstribution version
related facts were used, which required specific files, or links to
files for distributions in the same "family", which will probably have
the same variables set.
This change adds searching for files based on the `os_family` fact,
allowing distributions that follow the same family rules to be
supported, without any changes to the codebase. It is still possible
that a specific distribution configuration overrides the default
behavior, as `os_family` has lower priority than `distribution`.
For example, distributions on the `RedHat` family, like Oracle Linux,
Alma Linux, and Rocky Linux, work withoutadding new files, or links to
files, to fill the `vars`.
Fix issue #573. Fix issue #523.
When loading variables in all ansible-freeipa roles, it is expected
that a file with these variables is present for each supported Linux
distribution, and then, based on the information about the distribution
provided by Ansible, the correct file is loaded.
Previously, only the facts `distribution` and dinstribution version
related facts were used, which required specific files, or links to
files for distributions in the same "family", which will probably have
the same variables set.
This change adds searching for files based on the `os_family` fact,
allowing distributions that follow the same family rules to be
supported, without any changes to the codebase. It is still possible
that a specific distribution configuration overrides the default
behavior, as `os_family` has lower priority than `distribution`.
For example, distributions on the `RedHat` family, like Oracle Linux,
Alma Linux, and Rocky Linux, work withoutadding new files, or links to
files, to fill the `vars`.
Fix issue #573. Fix issue #523.
Some YAML files used in the project are not Ansible playbooks and
should not be evaluated by ansible-lint. This change add the paths
and files that should not be evaluated to an exclude list, that
affects linter operations in CI and pre-commit scripts.
There are common parameters in all modules like ipaadmin_principal and
ipaadmin_password. As this list of common parameters will be extended
soon, there is a need to reduce the code and documentation duplicates.
A ModuleDocFragment is added to provide the module documentation for the
common parameters. This is used in the modules with
extends_documentation_fragment.
ansible_freeipa_module has additional ipamodule_base_spec and
get_ipamodule_base_vars. ipamodule_base_spec extends argument_spec in
the module and get_ipamodule_base_vars is used to return a dict
containing the common parameters.
Use gen_add_list and gen_intersection_list for host, hostgroup, user,
group, allow_sudocmd, allow_sudocmdgroup, deny_sudocmd, deny_sudocmdgroup,
sudooption, runasuser, runasgroup and res_find member handling.
The functions are used to reduce the add lists to the new entries
only and the delete lists to the entries that are in the user and
the show list result.
This enables to remove the ignores for "already a member" and
"not a member" errors..
Single hostnames can be used for hbacrule_add_host and will match fqdn
in IPA internally. Simple host names have to be extended to be FQDN to
be able to compare them for _host_add and _host_remove.
Two new functions have been added to ansible_freeipa_module:
- api_get_domain - Get the domain from the api
- ensure_fqdn - Extend a single name with the domain
This fixes#617 - hbacrule_add_host: already a member
json_query is not part of Ansible Core. As this is the only used module
that is not in Core it has been replaced by constructig a new dict with
the needed user names from users dict using a loop and set_fact.
There is a new automount location management module placed in the plugins folder:
plugins/modules/ipaautomountlocation.py
This module allows to ensure presence and absence of automount locations
that act as containers for automount onjects in ipa.
Here is the documentation for the module:
README-automountlocation.md
New example playbooks have been added:
playbooks/automount/automount-location-absent.yml
playbooks/automount/automount-location-present.yml
New tests for the module:
tests/automount/test_automountlocation.yml
The automember_add_condition and automember_remove_condition commands
are not verifying condition keys in all cases. This is for example not
done in the removal case if a condition is not part of the automember
rule.
For consistent behaviour this needs to be done in the automember module
now. The condition keys are verified with the user and group aciattrs
returned by the API command json_metadata now.
Related: RHBZ#1976926
result["failed"] is used only for INCLUDE_RE, EXCLUDE_RE if entries could
not be added that are already there and it entries could not be removed
that are not there.
All other issues like invalid attributes etc. are handled as exceptions.
Therefore the error section is not here as in other modules.
Fixes RHBZ#1979626
Instead of ignoring errors it is better that we specific evaluate
errors provided by negative tests. This patche removes the an
occurrence of 'ignore_errors: true' in ansible-freeipa tests.
The parameters in the inclusive and exclusive dicts are not defined and
also not checked. Therefore it is possible to have parameters in the
dicts that are not used.
The dicts for the inclusive and exclusive parameters are now defined and
also added to the description.
Ansible 2.11 does not like to have empty dependencies. It will fail to
install the collection with "'NoneType' object has no attribute 'items'"
error. The empty dependencies have been removed.
The requirements for ansible in requirements.txt have been removed also.
Fixes#571 (error installing freeipa-ansible_freeipa-0.3.6.tar.gz)
Up to now a python snippet was used to get IPA_BACKUP_DIR from ipaplatform
but this was not working when ansible_facts was false due to not getting
ansible_python_interpreter set.
The module version is also working if gather_facts is turned off.
This patch modify Github 'lint' workflow to execute each linter
verifications as a separate job. This will allow us to easily see
which linter has failed, and ensure that all are executed, even
if one fails.
