Now that we don't need to worry about python-2.4 and 2.5, we can make
some improvements to the way AnsiballZ handles modules.
* Change AnsiballZ wrapper to use import to invoke the module
We need the module to think of itself as a script because it could be
coded as:
main()
or as:
if __name__ == '__main__':
main()
Or even as:
if __name__ == '__main__':
random_function_name()
A script will invoke all of those. Prior to this change, we invoked
a second Python interpreter on the module so that it really was
a script. However, this means that we have to run python twice (once
for the AnsiballZ wrapper and once for the module). This change makes
the module think that it is a script (because __name__ in the module ==
'__main__') but it's actually being invoked by us importing the module
code.
There's three ways we've come up to do this.
* The most elegant is to use zipimporter and tell the import mechanism
that the module being loaded is __main__:
* 5959f11c9d/lib/ansible/executor/module_common.py (L175)
* zipimporter is nice because we do not have to extract the module from
the zip file and save it to the disk when we do that. The import
machinery does it all for us.
* The drawback is that modules do not have a __file__ which points
to a real file when they do this. Modules could be using __file__
to for a variety of reasons, most of those probably have
replacements (the most common one is to find a writable directory
for temporary files. AnsibleModule.tmpdir should be used instead)
We can monkeypatch __file__ in fom AnsibleModule initialization
but that's kind of gross. There's no way I can see to do this
from the wrapper.
* Next, there's imp.load_module():
* https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151
* imp has the nice property of allowing us to set __name__ to
__main__ without changing the name of the file itself
* We also don't have to do anything special to set __file__ for
backwards compatibility (although the reason for that is the
drawback):
* Its drawback is that it requires the file to exist on disk so we
have to explicitly extract it from the zipfile and save it to
a temporary file
* The last choice is to use exec to execute the module:
* https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175
* The code we would have to maintain for this looks pretty clean.
In the wrapper we create a ModuleType, set __file__ on it, read
the module's contents in from the zip file and then exec it.
* Drawbacks: We still have to explicitly extract the file's contents
from the zip archive instead of letting python's import mechanism
handle it.
* Exec also has hidden performance issues and breaks certain
assumptions that modules could be making about their own code:
http://lucumr.pocoo.org/2011/2/1/exec-in-python/
Our plan is to use imp.load_module() for now, deprecate the use of
__file__ in modules, and switch to zipimport once the deprecation
period for __file__ is over (without monkeypatching a fake __file__ in
via AnsibleModule).
* Rename the name of the AnsiBallZ wrapped module
This makes it obvious that the wrapped module isn't the module file that
we distribute. It's part of trying to mitigate the fact that the module
is now named __main)).py in tracebacks.
* Shield all wrapper symbols inside of a function
With the new import code, all symbols in the wrapper become visible in
the module. To mitigate the chance of collisions, move most symbols
into a toplevel function. The only symbols left in the global namespace
are now _ANSIBALLZ_WRAPPER and _ansiballz_main.
revised porting guide entry
Integrate code coverage collection into AnsiballZ.
ci_coverage
ci_complete
* aws_eks: New module for managing AWS EKS
aws_eks module is used for creating and removing EKS clusters.
Includes full test suite and updates to IAM policies to enable it.
* Clean up all security groups
* appease shippable
* Rename aws_eks module to aws_eks_cluster
The compute policy was exceeding maximum size and contained
policies that already exist in ecs-policy.
Look up suitable AMIs rather than hardcode
We don't want to maintain multiple image IDs for multiple regions
so use ec2_ami_facts to set a suitable image ID
Improve exception handling
* Fix test-module failing to validate args
The test-module pass a wrong argument _ansible_tmp cause the validation failed.
Change the argument _ansible_tmp to _ansible_tmpdir to fix this.
* Add a integration test for test-module.
Prior to this change, we don't have a test for test-module.
This change ensure the correctness of test-module script.
Remove VPC permissions from network-policy.json as they mostly duplicate
compute-policy.json permissions - separating the VPC and compute permissions
would likely lead to further confusion.
We only needed it for migrating cherrypicks between the unified repo and
the ansible-modules-* repos. Now that we aren't supporting 2.3, we no
longer need this script.
* New module: ec2_vpc_vpn_facts
* Add integration tests for ec2_vpc_vpn_facts and the IAM permissions
* Add retry to VPC removal
* Use unique name for VGW
* Always clean up after tests and add retries
* create module tmpdir based on remote_tmp
* Source remote_tmp from controller if possible
* Fixed sanity test and not use lambda
* Added expansion of env vars to the remote tmp
* Fixed sanity issues
* Added note around shell remote_tmp option
* Changed fallback tmp dir to ~/.ansible/tmp to make shell defaults
* Add helpful failure message if target_type=ip is not supported
Create test case for target_type=ip not supported
* Update elb_target_group module to latest standards
Use AnsibleAWSModule
Improve exception handling
Improve connection handling
Enable awsvpc network mode for ECS services and tasks and
their underlying task definitions
Improve test suite to thoroughly test the changes
Use runme.sh technique to run old and new versions of botocore to
ensure that the modules work with older botocore and older network modes
and fail gracefully if awsvpc network mode is used with older botocore
* Add aws_ses_identity_policy module for managing SES sending policies
* Add option to AnsibleAWSModule for applying a retry decorator to all calls.
* Add per-callsite opt in to retry behaviours in AnsibleAWSModule
* Update aws_ses_identity_policy module to opt in to retries at all callsites.
* Add test for aws_ses_identity_policy module with inline policy.
* Remove implicit retrys on boto resources since they're not working yet.
Cloudfront needs CreateOriginAccessIdentity
Add profile parameter to setup-iam.yml. Could arguably just use
AWS_PROFILE but given that other tasks are using profile, should
be consistent.
* 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
* Add aws_ses_identity module
* Update CI alias, add BotoCoreError exception handling.
* Add SES and SNS permissions to hacking/aws_config to run aws_ses_identity integration tests
The search string used to look for Clear Linux
was changed in 45a9f96774 to
be more specific, but was too specific. Now finding
a substring match for 'Clear Linux' in /usr/lib/os-release
is enough to consider a match.
Since the details of the full name in os-release varies
('Clear Linux Software for Intel Architecture',
'Clear Linux OS for Intel Architecture', etc) the
search string match was failing and would fall back to the
'first word in the release file' method resulting in
ansible_distribution='NAME="Clear'
Also add a meta fact indicating which search string
was matched.
Test case info from:
https://github.com/ansible/ansible/issues/31501#issuecomment-340861535Fixes#31501
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
* added cloudfont.py, modified cloudfront_facts.py class name and fixed a minor bug
* Improvements to cloudfront_distribution
* Reduce the scope of the cloudfront_distribution module
* Remove presigning
* Remove streaming distribution functionality
* Add full test suite for cloudfront distribution
* Meet Ansible AWS guidelines
* Make requested changes
Fix tests
Use built-in waiter
Update copyright
Tests for:
* ecs_cluster
* ecs_service
* ecs_service_facts
* ecs_taskdefinition
* ecs_taskdefinition_facts
* Add idempotency testing
Test ecs_cluster, ecs_service and ecs_taskdefinition for trivial
idempotency. Add FIXMEs to the tests because the latter two fail.
Remove unused dependencies
This fix adds replacement for exit() to sys.exit(), as
exit() is not recommended way to exit from the program.
Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>
* Add some integration tests for ec2_vpc_net module
* Add a couple tests for check mode
fix typo
ensure the DHCP option set is cleaned up
* Add permissions to test policy
* 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
Exception was:
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 155, in boilerplate_module
if module_style == 'new' and 'ANSIBALLZ_WRAPPER = True' in module_data:
TypeError: a bytes-like object is required, not 'str'
* Add integration test suite for ec2_vpc_subnet
* wrap boto3 connection in try/except
update module documentation and add RETURN docs
add IPv6 support to VPC subnet module
rename ipv6cidr to ipv6_cidr, use required_if for parameter testing, update some failure messages to be more descriptive
DryRun mode was removed from this function a while ago but exception handling was still checking for it, removed
add wait and timeout for subnet creation process
fixup the ipv6 cidr disassociation logic a bit per review
update RETURN values per review
added module parameter check
removed DryRun parameter from boto3 call since it would always be false here
fix subnet wait loop
add a purge_tags parameter, fix the ensure_tags function, update to use compare_aws_tags func
fix tags type error per review
remove **kwargs use in create_subnet function per review
* rebased on #31870, fixed merge conflicts, and updated error messages
* fixes to pass tests
* add test for failure on invalid ipv6 block and update tags test for purge_tags=true function
* fix pylint issue
* fix exception handling error when run with python3
* add ipv6 tests and fix module code
* Add permissions to hacking/aws_config/testing_policies/ec2-policy.json for adding IPv6 cidr blocks to VPC and subnets
* fix type in tests and update assert conditional to check entire returned value
* add AWS_SESSION_TOKEN into environment for aws cli commands to work in CI
* remove key and value options from call to boto3_tag_list_to_ansible_dict
* remove wait loop and use boto3 EC2 waiter
* remove unused register: result vars
* revert az argument default value to original setting default=None
* Allow backoff for describe_subnets
Improve exception handling to latest standards
* Add integration test suite for ec2_vpc_subnet
* Add test for creating subnet without AZ
Fix bug identified by test
Fixes#31905
* ec2_group: add support for rule descriptions.
* Document rule description feature and add an example using it.
* Fix removing rule descriptions.
* Add integration tests to verify adding/modifying/removing rule descriptions works as expected.
* Add permissions to hacking/aws_config/testing_policies/ec2-policy.json for updating ingress and egress rule descriptions.
* ec2_group: add backwards compatibility with older versions of botocore for rule descriptions.
* Add compatibility with older version of botocore for ec2_group integration tests.
* ec2_group: move HAS_RULE_DESCRIPTION to be checked first.
* Make requested change
* Pass around a variable instead of client
* Make sure has_rule_description defaults to None
* Fail if rule_desc is in any ingress/egress rules and the the botocore version < 1.7.2
* Remove unnecessary variable
* Fix indentation for changed=True when updating rule descriptions.
* minor refactor to remove duplicate code
* add missing parameter
* Fix pep8
* Update test policy.
Provide all necessary permissions for AMI tests
Allow tests to run in us-east-2
Ensure `always` section gets used
Update tests to ensure that cleanup works better, and add
deletion idempotency test
After running hacking/test-module to generate some output,
the JSON output can be fed into the return skeletion generator
to create an excellent starting point for RETURN docs
* Update RDS parameter group for boto3
* Update to boto3
* Update to latest ansible standards
* Remove choices list for valid engines (See #19221 for context)
* Allow tagging
* Return some useful information, and document that information
* Add tests for rds_param_group
* Improve testing of rds_param_group
* Add purge_tags option for rds_param_group
* Fix remaining broken rds_param_group tests
* Ensure the group name is lowercased. Fixes integration tests when run on OSX
* Add network value to support_by field.
* New support_by value, certified
* Deprecate curated in favor of certified
* Add conversion from 1.0 to 1.1 to metadata-tool
* Add supported by Red Hat field to ansible-doc output
* Add test runner to PATH
* Add Python3 support
* Updating env-setup.fish to use more portable '-exec' rather than '-delete'
* Create gen_egg_info function
Move code into a function similar to env-setup.
Silence all output when run with -q
Toshio Kuratomi