When we read files from disk we always want to read them as bytes and
then convert them to text ourselves. This gives us greater control over
what encodings are used, what to do in case of errors decoding the bytes
to text, and better resilience against problems on both Python 2 and
Python 3.
If we left it up to Python to do this, on Python2, this could mean
that config values end up as bytes (leading to tracebacks elsewhere in
the code). In Python3, it could traceback if the user's locale did not
match with the encoding of the ini file or config files could be decoded
as the user's locale encoding instead of as utf-8.
* VMware: Allow user to select disk_mode
This fix allows user to select disk modes for given disk configuration
in the given VM.
Fixes: #37749
Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>
* Review comments
Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>
* Don't ignore a duplicate host for an already processed include, assume that the repetition indicates a new include. Fixes#40317
* Add intg tests to ensure duplicate items in loop are not deduped
* Add note about relative indexing
* add example using peer keepalive vrf and delay restore
<!--- Your description here -->
add example using peer keepalive vrf and delay restore
+label: docsite_pr
+label: issue ansible/community#311
* Update nxos_vpc.py
update task name to include "existing"
The LDAP values may be of any kind (pictures, bytes, etc.) thus, ldap module enforce a "bytes" type.
We should pass properly encoded values instead of str
Fixes: #39569
Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>
* Test case for missing permissions
* Update aws_s3 module to latest standards
* Use AnsibleAWSModule
* Handle BotoCoreErrors properly
* Test for BotoCoreErrors
* Check for XNotImplemented exceptions (#38569)
* Don't prematurely fail if user does not have s3:GetObject permission
* Allow S3 drop-ins to ignore put_object_acl and put_bucket_acl
* If the command input is dict from module
in that case the check to see if command is
end or `!` to exclude it from executing on remote
host is wrong. Fix the logic to check `end` and
`!` commands
* 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
* changing metadata to core support
* changing metadata to core support
* changing metadata to core support
* changing metadata to core support
* changing metadata to core support
* Changed request() to run json.loads() instead of module doing it
- Removed json.loads() from modules
- Removed some unreliable integration tests
- Removed self.function setting in construct_path()
-
* PEP8 changes
* Remove debug line for push
* Allow the use of 'aws:kms' as an encryption method
* Allow the use of a non standard KMS key
* Deduce whether AWS Signature Version 4 is required rather than specifying with a parameter
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
Fargate instances do not require memory and cpu descriptors. EC2 instances
do require descriptions. https://botocore.readthedocs.io/en/latest/reference/services/ecs.html#ECS.Client.describe_task_definition
Fargate requires that cpu and memory be defined at task definition level.
EC2 launch requires them to be defined at the container level.
Fargate requires the use of awsvpc for the networking_mode. Also updated,
the documentation regarding where and when memory/cpu needs to the assigned.
The task_definition variable for the awspvc configuration colided with
the ecs_service for the bridge network. This would cause the test to fail.
Add testing for fargate
Add examples for fargate and ec2
* Adding a switch to print logs
* Update to docs
* Adding `version_added` for the new option
* Updating version added
* Moving `output_logs` before `paued` for alphabetically ordering.
When using an empty string as the version argument, the module would
before attempt to run something akin to:
pip install module==""
This changes the behavior to:
pip install module
Fixes#41043
* Add support for classmap-authoritative option for composer
* Add support for apcu-autoloader option for composer
* Fix a typo for the no-plugins option
* Refactor ios cliconf plugin and ios_config module
* Refactor ios cliconf plugin to support generic network_config module
* Refactor ios_config module to work with cliconf api's
* Enable command and response logging in cliconf pulgin
* cliconf api documentation
* Fix unit test and other minor changes
* Doc update
* Fix CI failure
* Add default flag related changes
* Minor changes
* redact input command logging by default
In the process of building up the inventory by parsing each inventory
source with each available inventory plugin, there are three kinds of
possible errors (listed in order from earliest to latest):
1. One source could not be parsed by a particular plugin.
2. One source could not be parsed by any available plugin.
3. ALL sources could not be parsed by any available plugin.
The errors in (1) are a part of normal operation, e.g., the script
plugin is expected to fail to parse an ini-format source, and we will
ignore that error and try the next plugin. There is currently no way to
control this, and no known compelling use-case for a setting to control
it. This commit does not make any changes here.
We implement "any_unparsed_is_failed" to handle (2) above. If enabled,
this requires that every available source be parsed validly by at least
one plugin. In an inventory comprising a static hosts file and ec2.py,
this setting will cause a fatal error if ec2.py fails (a situation that
attracted only a warning earlier).
We clarify that the existing "unparsed_is_failed=true" setting causes a
fatal error only in (3) above, i.e., if NO inventory source could be
parsed. In other words, if there is ANY valid source in the inventory
(e.g., an ini-format static file), no combination of errors and the
setting will cause a fatal error.
If you want to execute your playbooks when your inventory is…
(a) complete, use "any_unparsed_is_failed=true".
(b) not empty, use "unparsed_is_failed=true".
The "unparsed_is_failed" setting should be renamed to
"all_unparsed_is_failed", but this commit does not do so.
Fixes#40512Fixes#40996
* Updating tower_job_template.py
* tower_job_template: Update parameter version_added to 2.7
* Ensure that unset credentials aren't passed
Passing empty strings for unset credentials causes ValueErrors as
the API expects an integer. Don't pass unset credentials
* tower_inventory_source: Add support for the inventory source via ansible-tower-cli.
* Add test coverage for tower_inventory_source.
* Update version_added to 2.7
* diff in as-path-set or prefix-set
* fix caveat diff can not have last line with comma in prefix-set/as-path/community-set
* Simplify fix to include indentation before parse
* remove debugger
* route-policy diffs
* fix iosxr_config crash issue
* new changes in iosxr_config after git add
* end-policy-map and end-class-map are properly indented so match misplaced children only when end-* is at the beigining also fix pep8
* Remaining config blocks of route-policy which needs exclusion from diff. added new tests
* pylint/pep8 warnings
* Review comments , sanity test fix
* shbang warning
* remove unused import
* added validate_certs option to zabbix_*_facts modules and fixed documentation to use doc fragment
* removed code duplication from zabbix_host_facts module
The parent parameter was not being used in the module. This meant
that all child profiles would use the system-defined parent instead
of the parent specified in the module
* fix ec2_ami block_device_mapping size to be int
* fixed cr issues
renamed `type` to `attribute_type`
reused `new_item` instead of creating new variable `value`
Fixes: #40631
The module was not correctly handling a certain state where the node
could be down, but the monitor was enabled on the node.
This patch fixes it.
With the addition on ajson.py in cbb6a7f4e8, two
new classes were created: AnsibleJSONDecoder and AnsibleJSONEncoder. These
classes are used when calling json.looads() and json.dumps().
This works fine with everything except the jsonfile.py cache plugin, which would
first try to import simplejson as json, then fall back to json. When simplejson
is installed, the load() or dump methods from simplejson are called, which then
try to use the AnsibleJSONEncoder/AnsibleJSONDecoder subclass from ajson.py.
But asjon.py imports json, not simplejson, and things blow up.
The debug message stated that the vault password client script was invoked as `example.py --vault-id=example`, while it's actually `example.py --vault-id example`.
* Adds requests.Session like class
* py2 syntax fix
* Add a few examples to the Request docstrings
* Add helper methods and docs
* Fix test failures
* Switch tests to test Request instead of open_url, add simple open_url test to validate funcitonality
* Fix filename in replace-urlopen code smell test
Removes the SSH enforcement because it breaks situations where, for example,
an admin maintains Ansible Tower and another team needs to use Ansible
with the BIG-IP modules, but the Tower admin does not want to install
the F5-SDK on the Ansible Tower instance.
In this situation, the user needs to delegate the execution of the BIG-IP
module to another linux host that does have the dependencies installed.
Usually this is done with delegation, often over SSH. Limiting the SSH
connection here prevents this from working.
