* document how to migrate between collections
* Apply suggestions from code review
Co-authored-by: John R Barker <john@johnrbarker.com>
Co-authored-by: Felix Fontein <felix@fontein.de>
Follow up to #70221
Related to #67794
CVE-2020-1736
When set_mode_if_different() is called with mode of 'None', ensure we issue
a warning about the change in default permissions.
Add integration tests to ensure the warning works properly.
* Fix tests
- actually use custom module 🤦♂️
- verify file permission on created files
- use remote_tmp_dir so we're ready for split controller
- improve test module so we can skip the call to set_fs_attributes_if_different()
- fix tests for CentOS 6
Change:
- Clarify that not passing `--tags` will cause `ansible_run_tags` to
default to `["all"]`.
- Add some extra coverage around `ansible_run_tags`
Test Plan:
- New integration and unit tests
Tickets:
- Fixes#69619
Signed-off-by: Rick Elrod <rick@elrod.me>
- Combine remote and local debugging instructions.
- Update the example code to match current AnsiballZ structure and behavior
- Change reference name and update references
- Clarify how PYTHON path is modified
- Also add note about other remote debugging tools.
Co-authored-by: Alicia Cozine <879121+acozine@users.noreply.github.com>
* Also validate top-level version_added.
* Fix error code.
* Produce same version_added validation error in schema than in code (and stop returning it twice).
* Return correct error codes for invalid version_added for options and return values.
* Add changelog.
* Fix forgotten closing braket.
* Accept 'historical' for some top-level version_added.
PR #70446: it's a follow-up for #70445.
It includes a merge of `examples/scripts/uptime.py` and a similar
code snippet from `docs/docsite/rst/dev_guide/developing_api.rst`.
This patch also changes the docs RST file to include contents of
the example file instead of holding a copy of a similar code.
* Change default file permissions so they are not world readable
CVE-2020-1736
Set the default permissions for files we create with atomic_move() to 0o0660. Track
which files we create that did not exist and warn if the module supports 'mode'
and it was not specified and the module did not call set_mode_if_different(). This allows the user to take action and specify a mode rather than using the defaults.
A code audit is needed to find all instances of modules that call atomic_move()
but do not call set_mode_if_different(). The findings need to be documented in
a changelog since we are not warning. Warning in those instances would be frustrating
to the user since they have no way to change the module code.
- use a set for storing list of created files
- just check the argument spac and params rather than using another property
- improve the warning message to include the default permissions
* Format using `` instead of `, add line breaks for long lines, rephrase or remove useless text.
Move some text.
* Add clearer version of OpenSSh is affected by SCP bug
* Review some pages using ansible doc writing guide.
From what I have observed it is the block itself that doesn't support the name attribute rather than the tasks inside the block.
* Update docs/docsite/rst/user_guide/playbooks_blocks.rst
Co-authored-by: Alicia Cozine <879121+acozine@users.noreply.github.com>
* Build documentation for Ansible-2.10 (formerly known as ACD).
Builds plugin docs from collections whose source is on galaxy
The new command downloads collections from galaxy, then finds the
plugins inside of them to get the documentation for those plugins.
* Update the python syntax checks
* docs builds can now require python 3.6+.
* Move plugin formatter code out to an external tool, antsibull-docs.
Collection owners want to be able to extract docs for their own
websites as well.
* The jinja2 filters, tests, and other support code have moved to antsibull
* Remove document_plugins as that has now been integrated into antsibull-docs
* Cleanup and bugfix to other build script code:
* The Commands class needed to have its metaclass set for abstractmethod
to work correctly
* Fix lint issues in some command plugins
* Add the docs/docsite/rst/collections to .gitignore as
everything in that directory will be generated so we don't want any of
it saved in the git repository
* gitignore the build dir and remove edit docs link on module pages
* Add docs/rst/collections as a directory to remove on make clean
* Split the collections docs from the main docs
* remove version and edit on github
* remove version banner for just collections
* clarify examples need collection keyword defined
* Remove references to plugin documentation locations that no longer exist.
* Perhaps the pages in plugins/*.rst should be deprecated
altogether and their content moved?
* If not, perhaps we want to rephrase and link into the collection
documentation?
* Or perhaps we want to link to the plugins which are present in
collections/ansible/builtin?
* Remove PYTHONPATH from the build-ansible calls
One of the design goals of the build-ansible.py script was for it to
automatically set its library path to include the checkout of ansible
and the library of code to implement itself. Because it automatically
includes the checkout of ansible, we don't need to set PYTHONPATH in
the Makefile any longer.
* Create a command to only build ansible-base plugin docs
* When building docs for devel, only build the ansible-base docs for
now. This is because antsibull needs support for building a "devel
tree" of docs. This can be changed once that is implemented
* When building docs for the sanity tests, only build the ansible-base
plugin docs for now. Those are the docs which are in this repo so
that seems appropriate for now.
A little further down the page is another index, ansible_loop.index, which shares a similar description but is 1 indexed.
Its zero indexed twin has a 0 suffix.
``ansible_loop.index`` The current iteration of the loop. (1 indexed)
``ansible_loop.index0`` The current iteration of the loop. (0 indexed)
To remove ambiguity around the usage of index_var, explicitly mention that this variable is 0 indexed.
Change:
Our handling of NetBSD virtualization facts led to facts that were just
plain incorrect. One example is reporting Xen even when the system is
running on something completely different (like KVM).
As stated by the reporter of #69352, NetBSD has a better sysctl setting
to use for this information, machdep.hypervisor.
This PR does the following:
- Try to use machdep.hypervisor sysctl value if the other sysctl values
we check don't end up with enough information to be useful
- Only look for /dev/xencons and assume Xen if nothing else works
(Really this should probably return 'unknown' since the file exists on
non-Xen systems and is not very useful).
- Add a few more patterns (Xen matches and also Hyper-V) to
VirtualSysctlDetectionMixin#detect_virt_product.
This change is slightly breaking:
- If the first two attempts at using sysctl worked before,
(machdep.dmi.system-product and machdep.dmi.system-vendor), they will
continue to work.
- For cases when those values didn't work, previously the existence of
/dev/xencons was checked, and if found, we reported 'xen' (even on
non-Xen systems when the file existed). After this PR, we try the
machdep.hypervisor sysctl key before still falling back to
/dev/xencons. This means that in some cases, we might go from
(wrongly) saying "xen" to giving a more accurate value such as "kvm"
or "Hyper-V".
Test Plan:
- Tested with local NetBSD VM and got 'kvm' instead of 'xen' back.
Tickets:
- Fixes#69352
Signed-off-by: Rick Elrod <rick@elrod.me>
* Add type for options in the sample module shown in Developing Modules, as this is required to have the tests being
green
* Remove duplicated strings: the same sentence is in "Python tips" and "Module security", keeping the
latter.
Previous version initialized the `TaskQueueManager` after calling
`Play.load()` while advertising a way to inject a custom library
location path. This caused the tasks loader not to find any custom
modules because it was triggered before the path was actually added
to the module loader.
This patch changes the order of the operations to ensure that the
customized `context.CLIARGS` actually influences things.
Resolves https://github.com/ansible/ansible/issues/69758.
If a local module has no documentation, the doc command will fail without any hints of what is wrong. Add another way to confirm the presence of a local module.
* Update docs/docsite/rst/dev_guide/developing_locally.rst
Co-authored-by: Abhijeet Kasurde <akasurde@redhat.com>
* Update docs/docsite/rst/dev_guide/developing_collections.rst
* add steps to create changelogs, add sentence about not using the tool
* add note for rerunning the command
Co-authored-by: Felix Fontein <felix@fontein.de>
* Update guide_gce.rst
The given example is not working as a SSH connection is not possible because of #59573. Instead, in this tutorial the default network should be used which allows SSH connections in most cases.
Co-authored-by: Sandra McCann <samccann@redhat.com>
* Updated docs/docsite/rst/dev_guide/developing_modules_documenting.rst, added documentation for R()
* added link to how to add anchors, anchor, and clarification on when to use each option
Co-authored-by: John R Barker <john@johnrbarker.com>
Co-authored-by: Alicia Cozine <879121+acozine@users.noreply.github.com>
* Remove community.crypto parts from porting guide.
* Remove community.general parts from porting guide.
* Removed 'modules removed' section, since this is better suited for a general entry on the collection split.
* Validate ansible-base & collection's runtime.yml
Add new test `runtime-metadata`
* Schema validation of file
* Error if a a legacy meta/routing.yml exist in a collection
* removal_date OR removal_version
* Add tombstone validation.
* Allow both ISO 8601 date strings and datetime.date objects (from YAML dates).
* Address review comments.
* Add metadata to test collection.
* Add requirements file.
Co-authored-by: Felix Fontein <felix@fontein.de>
Co-authored-by: Matt Clay <matt@mystile.com>
Replace the ansible-base changelog linting and generation tool with antsibull-changelog and make it available for linting collections. Previously changelog linting was limited to ansible-base.
* Allow to specify collection_name separately for deprecation.
* Use new functionality in Ansible.
* Use new functionality in tests.
* Update tagging/untagging functions.
* Update pylint deprecated sanity test.
* Update validate-modules. Missing are basic checks for version_added (validate semantic version format for collections).
* Improve version validation. Re-add version_added validation.
* Make sure collection names are added to return docs before schema validation.
* Extra checks to avoid crashes on bad data.
* Make C# module utils code work, and update/extend tests.
* Add changelog fragment.
* Stop extracting collection name from potentially tagged versions/dates.
* Simplify C# code.
* Update Windows modules docs.
* Forgot semicolons.
* Make AnsibleVaultEncryptedUnicode work more like a string. Fixes#24425
* Remove debugging
* Wrap some things
* Reduce diff
* data should always result in text
* add tests
* Don't just copy and paste, kids
* Add eq and ne back
* Go full UserString copy/paste
* Various version related fixes
* Remove trailing newline
* py2v3
* Add a test that can evaluate whether a variable is vault encrypted
* map was introduces in jinja2 2.7
* moar jinja
* type fix
Co-Authored-By: Sam Doran <sdoran@redhat.com>
* Remove duplicate __hash__
* Fix typo
* Add changelog fragment
* ci_complete
Co-authored-by: Sam Doran <sdoran@redhat.com>
* starting metadata sunset
- purged metadata from any requirements
- fix indent in generic handler for yaml content (whey metadata display was off)
- make more resilient against bad formed docs
- removed all metadata from docs template
- remove metadata from schemas
- removed mdata tests and from unrelated tests
Co-authored-by: Felix Fontein <felix@fontein.de>
Co-authored-by: Rick Elrod <rick@elrod.me>
Change:
On OpenBSD when using pipelining, we do not set cwd which results in a
permissions fatal. Ensure that `''` - cwd - is not in `sys.path`.
Test Plan:
Tested against local OpenBSD VM
Tickets:
Fixes#69320
Signed-off-by: Rick Elrod <rick@elrod.me>
* Enable installing collections from git repositories
* Add tests for installing individual and multiple collections from git repositories
* Test to make sure recursive dependencies with different syntax are deduplicated
* Add documentation
* add a changelog
* Skip Python 2.6
* Only fail if no collections are located in a git repository
Add support for a 'type' key for collections in requirement.yml files.
Update the changelog and document the supported keys and allowed values for the type.
Add a note that the collection(s) in the repo must contain a galaxy.yml
* Add a warning about embedding credentials in SCM URLs
* Update with review suggestions
* suppress sanity compile failure for Python 2.6
* Track collection for version_added.
Validate *all* version numbers in validate-modules.
For tagged version numbers (i.e. version_added), consider source collection to chose validation.
* Make tagging/untagging functions more flexible.
* Tag all versions in doc fragments.
* Tag all deprecation versions issued by code.
* Make Display.deprecated() understand tagged versions.
* Extend validation to enforce tagged version numbers.
* Tag versions in tests.
* Lint and fix test.
* Mention collection name in collection loader's deprecation/removal messages.
* Fix error IDs.
* Handle tagged dates in Display.deprecated().
* Also require that removed_at_date and deprecated_aliases.date are tagged.
* Also automatically tag/untag removed_at_date; fix sanity module removal version check.
* Improve error message when invalid version number is used (like '2.14' in collections).
* Improve the changelog-how-to documentation:
* remove repetitions
* mention starting with lowercase
* update the link to examples, remove an irrelevant link
* change win_ example to apt_repository
* add info about collections
* note that environment is not good for secrets
* Update docs/docsite/rst/user_guide/playbooks_environment.rst
Co-authored-by: Sam Doran <sdoran@redhat.com>
Sandra McCann