This change makes os_stack module idempotent. Otherwise, re-use of the
module fails with:
Error updating stack: ERROR: The Parameter (...) was not provided.
Fixes#3165.
* Moved JSON-RPC client IPAClient class to ansible.module_utils.ipa, which is extended by all ipa modules
* ipa_user: incorporate displayname and userpassword attributes in module_user
* ipa_user: capitalized "I" in comment
* ipa_user: updated get_ssh_key_fingerprint to include possibility of the uploaded SSH key including user@hostname comment, which also appears in the queried fingerprint. This fixes a mismatch in the calculated and queried SSH key fingerprint in the user_diff calculation when the user already exists.
* ipa_hbacrule: ipaenabledflag must be 'TRUE' or 'FALSE', not 'NO'
* ipa_sudorule: ipaenabledflag must be 'TRUE' or 'FALSE', not 'NO'
* Add author to files missing it
* Fix kibana
* More native YAML
* More native YAML
* More native YAML
* More native YAML. Now only languages/ is missing
* Use native yaml sintax for packaging/languages as well
* Some more and quote fixes
* Fix wrong grouping
The policycoreutils python API for RHEL6 and RHEL7 are sufficiently
different, requiring some additional definitions and specific conversion
that works on old and new implementations.
It also implements a fix for non-ascii error messages (like when using a
French locale configuration).
This fixes#3551.
* Added new option to select the active a10 partition
* added version_added to the description of the new option
* added RETURN documentation
* fixed indents
* Removed empty cases, removed unneeded aliases
* removed artifacts from merging
* updated version_added to 2.3
* removed host, username and password option
* removed write_config and validate_certs documentation
* add a10_server_axapi3 module
* added return documentation
* modified a10_server_axapi3.py per feedback
* fixed line 60 s/action/operation/
* modified a10_server_axapi3.py per feedback
* modified a10_server_axapi3.py per feedback
* corrected YAML format error in documentation
* removed slp_server_ip and slp_server check in code since the arguments are labeled as required, per feedback
* modified: a10_server.py
modified: a10_service_group.py
modified: a10_virtual_server.py
Changed main() block, restricted import to only functions used.
* removed space for main() to be last line
* removed invalid lines
* Modified Documentations for a10_server.py, a10_service_group.py, a10_virtual_server.py
* Take out alias:[] and choices:[] in Documentation from a10_service_group.py and a10_virtual_server.py since they are now the default
* deleted a10_server.py, a10_service_group.py, a10_virtual_server.py
* deleted 'version_last_modified' line in Documentation across a10_server.py, a10_service_group.py and a10_virtual_server.py as they were added in error, change validate_certs version_added in a10_server.py
* added newline after main()
* added newline after main() for a10_server_axapi3.py
* Create `serverless` module for handling deploys on the Serverless Framework
* fix interpreter line
* Successfully exit when a stage is already absent
Both the `homebrew` and `homebrew_cask` modules iterate over
dictionaries using `iteritems`. This is a Python 2-specific method whose
behavior is similar to `items` in Python 3+. The `iteritems` function in
the six library was designed to make it possible to use the correct
method.
Traceback:
Traceback (most recent call last):
File \"/tmp/ansible_d28_6uwl/ansible_module_make.py\", line 153, in <module>
main()
File \"/tmp/ansible_d28_6uwl/ansible_module_make.py\", line 119, in main
rc, out, err = run_command(base_command + ['--question'], module, check_rc=False)
File \"/tmp/ansible_d28_6uwl/ansible_module_make.py\", line 79, in run_command
return rc, sanitize_output(out), sanitize_output(err)
File \"/tmp/ansible_d28_6uwl/ansible_module_make.py\", line 95, in sanitize_output
return output.rstrip(b(\"\\r\\n\"))
TypeError: rstrip arg must be None or str
There is also a six.iteritems issue, fixed using six.
* gluster_volume: Fixes issue when creating a new volume failing due to peers not being present. The peers which are not 'localhost' should invoke wait_for_peer, but the find method returns -1 (not 0) on non-localhost peers.
By default, sparse option is true in ovirt. So the raw disk
creation in a block storage domain will fail with error "Disk
configuration (RAW Sparse) is incompatible with the storage domain
type".
The commit adds sparse option where it is send as False when
format is raw and True when format is qcow2
* New module proxmox_kvm
* fixed qxl value vor vga param
> | Name | Type | Format | Description |
> |------|------|--------|-------------|
> | vga | enum | std \| cirrus \| vmware \| qxl \| serial0 \| serial1 \| serial2 \| serial3 \| qxl2 \| qxl3 \| qxl4 | Select the VGA type. If you want to use high resolution modes (>= 1280x1024x16) then you should use the options 'std' or 'vmware'. Default is 'std' for win8/win7/w2k8, and 'cirrus' for other OS types. The 'qxl' option enables the SPICE display sever. For win* OS you can select how many independent displays you want, Linux guests can add displays them self. You can also run without any graphic card, using a serial device as terminal. |
* Fix create_vm() fail on PV 4.3
* Set default for force as null in doc
* proxmox_kvm: revision fixes
* proxmox_kvm: more revision fixes
* Fix indentation
* revision fixes
* Ensure PEP-3110: Catching Exceptions
* KeyError, to KeyError as -- PEP-3110: Catching Exceptions
* Fix Yaml document syntax; Notes: => Notes -
* Refix documentation issue
* Fix Documentation
* Remove Notes: in description
* Add current state and it return value
* Update documentation
* fixed local variable 'results' referenced before assignment
* Fix fixed local variable 'results' referenced before assignment
* minor fixes in error messages
* merge upstream/devel int devel
* minor fixes in error messages
* Fix indentation and documentation
* Update validate_certs description
* translate() has a different api for text vs byte strings
* maketrans must be imported from a different location on py2 vs py3
Since this is such a small string outside of a loop we don't have to
worry too much about speed so it's better to have a single piece of code
that works on both py2 and py3
Fixes#3249
The python-consul library already supports this, so it is just a simple
case of enablement.
This does not break the current logic in `add` of parsing as a check,
then parsing as a service if that fails… because service_name is
mandatory on a service registration and is invalid on a check
registration.
* ipa_group: Fix: 'list' object has no attribute 'get'
* ipa_hbacrule: Fix: 'list' object has no attribute 'get'
* ipa_host: Fix: 'list' object has no attribute 'get'
* ipa_hostgroup: Fix: 'list' object has no attribute 'get'
* ipa_role: Fix: 'list' object has no attribute 'get'
* ipa_sudocmd: Fix: 'list' object has no attribute 'get'
* ipa_sudocmdgroup: Fix: 'list' object has no attribute 'get'
* ipa_sudorule: Fix: 'list' object has no attribute 'get'
* ipa_user: Fix: 'list' object has no attribute 'get'
* ipa_sudorule: Fix: invalid 'cn': Only one value is allowed
* ipa_hostgroup: module returns changed if assigned hosts or hostgroups are not in lowercase
* extra detail on which step triggered 'change', detect and handle powershell mishandling nssm's unicode as utf8
* Simpler handling of nssm output encoding
Thanks @nitzmahone for a cleaner way to control PowerShell's behavior
While I still have no idea why or how the `map` call is being swapped out while still running in python 2.7, this change will fix the following error, as well as improve py3 compatibility.
* Add FreeIPA modules
* Update version_added from 2.2 to 2.3
* ipa_*: Use Python 2.4 syntax to concatenate strings
* ipa_*: Replace 'except Exception as e' with 'e = get_exception()'
* ipa_*: import simplejson if json can't be imported
* ipa_hbacrule: Fix: 'SyntaxError' on Python 2.4
* ipa_sudorule: Fix: 'SyntaxError' on Python 2.4
* ipa_*: Fix 'SyntaxError' on Python 2.4
* ipa_*: Import get_exception from ansible.module_utils.pycompat24
* Add FreeIPA modules
* Update version_added from 2.2 to 2.3
* ipa_*: Fix 'SyntaxError' on Python 2.4
* ipa_*: Replace Python requests by ansible.module_utils.url
* ipa_*: Replace Python requests by ansible.module_utils.url
* ipa_*: Add option validate_certs
* ipa_*: Remove requests from Ansible module documentation requirements
* ipa_sudorule: Remove unnecessary empty line
* ipa_sudorule: Remove markdown code from example
* ipa_group: Add choices of state option
* ipa_host: Rename options nshostlocation to ns_host_location, nshardwareplatform to ns_hardware_platform, nsosversion to ns_os_version, macaddress to mac_address and usercertificate to user_certificate and add aliases to be backward compatible
* Return actual queue attributes with result
Previously this was only returning the desired queue attributes, and not even returning the QueueARN for use elsewhere. Now it will return "results.attributes" that is retrieved with boto's get_queue_attributes().
* update return structure to reflect current SQS config; add documentation of return values
* Remove redundancy from if/else statement
Added support to explicitly manage task definitions be revision. If the
revision expectations of the ansible task cannot be met, an error is
thrown.
If revision is not explicitly specified, enhanced module to be
idempotent with respect to task definitions. It will search for an
active revision of the task definition that matches the containers and
volumes specified. If none can be found, a new revision will be created.
Currently <active> tag is passed within the disk element which is
incorrect. As a result, disk will remain inactive even though the
default option is true.
In Zabbix, the visible name defaults to the hostname. This is not very
useful if you try to manage vmware VMs as the so called host_name within
zabbix must be set to the vcenter UUID.
This patch allows you to provide an alias which will be shown with
zabbix. If its not supplied it will default to host_name.
* Adding support for proxysql
* Moved and restricted imports, updated exception handling
* Updated version_added, and mysqldb_found constant name
* Removed unnecessary parentheses
In dnf, environment groups are separate from groups. Need to handle
them separately when calling the API.
Fixes#2178
After upstream review, hande dnf-2.0 mandatory packages in groups
If mandatory packages in a group are not installed, a group will report
failure. Fix this by catching the error and reporting after trying
to install the other packages and groups in the transaction.
* dnf: fix compatibility with DNF 2.0
* Reimplement (copy) old dnf.cli.commands.parse_spec_group_file(),
upstream uses argparse since 2.0.
* add_remote_rpm() has been changed to the add_remote_rpms()
Closes: https://github.com/ansible/ansible-modules-extras/issues/3310
Signed-off-by: Igor Gnatenko <ignatenko@redhat.com>
* fixup! dnf: fix compatibility with DNF 2.0
This module aims to allow a user to manage the lifecycle of OpenSSL
private keys. Internally it relies on the pyOpenSSL python library
to interact with openssl.
A user is able to specify :
* key size (via `size` parameter)
* key algorithm (via `type` parameter)
* key location (via `path` parameter)
The most simple use case is:
```
- name: Generate ansible.com.pem SSL private key
openssl_privatekey: name=ansible.com.pem
path=/etc/ssl/private
```
A user can speficy more settings:
```
- name: Generate ansible.com.pem SSL private key
openssl_privatekey: name=ansible.com.pem
path=/etc/ssl/private
size=2048
type=DSA
```
A user can also force the regeneration of an SSL key:
```
- name: Generate ansible.com.pem SSL private key
openssl_privatekey: name=ansible.com.pem
path=/etc/ssl/private
force=true
```
* Changed default deployment mode to match with azure -cli behaviour. "Complete" mode by default is too dangerous.
* Set incremental as default behaviour for deployment mode.
This commit adds module settings for configuring TLS encyption on the
mqtt notification module. Previously there was no way to configure
sending the messages encrpyted to mqtt.
* Add initial support for using a cluster instead of an esxi hostname
* FindByInventoryPath doesn't always work, so implement a fallback method to match the path
Traceback (most recent call last):
File \"/tmp/ansible_ueg52c0b/ansible_module_blockinfile.py\", line 319, in <module>
main()
File \"/tmp/ansible_ueg52c0b/ansible_module_blockinfile.py\", line 259, in main
if line.startswith(marker0):
TypeError: startswith first arg must be bytes or a tuple of bytes, not str
Also clean imports while on it.
When using the `make:` extras module, often the action taken
by the Make target is large. It is useful, therefore, to see
the output that `make` had to std{out,err} during execution
even when the target did not fail.
Signed-off-by: Steve Kuznetsov <skuznets@redhat.com>
Fail with:
Traceback (most recent call last):
File \"/tmp/ansible_2rsl4fg1/ansible_module_pkgng.py\", line 353, in <module>
main()
File \"/tmp/ansible_2rsl4fg1/ansible_module_pkgng.py\", line 330, in main
_changed, _msg = install_packages(module, pkgng_path, pkgs, p[\"cached\"], p[\"pkgsite\"], dir_arg)
File \"/tmp/ansible_2rsl4fg1/ansible_module_pkgng.py\", line 161, in install_packages
old_pkgng = pkgng_older_than(module, pkgng_path, [1, 1, 4])
File \"/tmp/ansible_2rsl4fg1/ansible_module_pkgng.py\", line 121, in pkgng_older_than
while compare_version[i] == version[i]:
TypeError: 'map' object is not subscriptable
Fix an issue with _nic_to_public_ips_instance() function. There was an assumption in the code that the Public IP sits in the same resource group, this isn't always the case.
ansible-validate-modules is now in ansible/ansible
During 2.3 we will be merge the modules into ansible/ansible so this file will go away.
The new testing documentation will refer to `ansible-test` which will wrap up the unit, integration, and ansible-validate-modules. So no need to document here.
* openbsd_pkg: Use correct part of name in match.
Previously this part of the code could assume that the name was a stem
with nothing else attached (like "autoconf"). With the introduction of the
branch syntax ("autoconf%2.13") this is no longer true.
Check if the package name was identified as using a "branch" style name,
and base the match on the leading part of the name if that is the case.
While here remove unnecessary "pass" and tidy up debug log message.
Problem reported by @jasperla.
* openbsd_pkg: Add missing "." in comment.
Fix mongodb_user.py version detection logic for mongo srv 2.6 and mongo driver 2.7. The wrong variable was used for detecting the mongo driver version. This fix resolves the error "(Note: you must use pymongo 2.7+ with MongoDB 2.6.. 2.6.11)" no matter what version of pymongo you had installed for mongodb 2.6.
When archiving multiple files, the full length of the calculated
`arcroot` would be removed from the beginning of all file names. If
there was no arcroot, the first two characters of all files would be
removed, so `file.txt` would become `le.txt`.
This patch uses the regular expressions substitution anchored to the
start of the string, so the arcroot is only removed if it is actually
present.
* Fix error in crypttab doc
* Use path type for file argument
This permit to make sure that $HOME and '~' are properly
expanded, even if in the case of crypttab, this might not make
a lot of sense
* PanOS module that allows admin password change.
* fixed a typo
* empty __init__.py
* added require ansible Python extension .py
* added version string
* added return docstring
* changed version to 2.2
* - changes to the format and module as requested by @privateip
* changed version back to 2.2 as 2.3 was failing automated tests
* Revert "changed version back to 2.2 as 2.3 was failing automated tests"
reverting version info
This reverts commit 71d520f3b4b69eb017c2b9f287a74cb77fae9d1c.
The modules listed in this PR were using YAML that resulted in
blockquote tages being inserted into the generated RestructedText.
This PR fixes that so that the documentation once again looks correct
The returns are actually nested under `configuration` keys, so the docs
need to reflect that. Also add the automatic return of the function
version, so it can be used to feed the `lambda_alias` module.
The fix allows the asa_config module to request the config to contain
all default statements or password information necessary for vpn
tunnel endpoints
This argument had a couple of issues with it. First, as it was
being interpreted in the code, it did not check for idempotency.
Second, the model of having the parameters be "all_*" is going to
hinder the ability to "undo", so-to-speak, what the user did while
maintaining legibility.
Consider if the user specified "all_enabled_vlans='net1'" and then
decided they wanted to backout of this decision. What is the proper
argument to fulfill this wish? "all_enabled_vlans='...?'
This patch changes the all_enabled_vlans argument to be "enabled_vlans",
ensures that idempotency works, and also provides for a way to "undo" a
change to the enabled VLANs by allowing the user to specify the special
case VLAN named "ALL" (all capitals).
This makes the parameter more intuitive because the users will specify
which vlans they want to make the virtual available on
* enabled_vlans="net1"
but also allows them to "undo" what they did by setting it back with
the case of all
* enabled_vlans="ALL"
* elasticsearch_plugin: rewrite module to not use unsupported parameters (#1785)
Avoid using parameters when they are not needed (#1785)
* elasticsearch_plugin: add version only during plugin installation and parse plugin name for its removal
* elasticsearch_plugin: join command args before running it
We were incorrectly making VLANS always be untagged when they could
be either tagged or untagged. This change corrects the arguments to
the vlan module to allow for specifying either untagged or tagged
interfaces. The arguments are mutually exclusive
The code for traffic groups was not being tested and therefore
had errors associated with it. It is now covered in coverage tests
and bugs that were found in it have been fixed.
See this issue for details
https://github.com/F5Networks/f5-ansible/issues/28
In the six package, the map() function returns an iterator instead
of a list. This code was continuing to use the map() return value
as if it were a list and this broke the address_class facts.
This patch changes the code to use the list() method on the return
value of map().
Using values caused problems while creating an integration playbook as it is a reserved word. Seeing as this module is not yet released, it's prudent to make this change now. 'record_data' is more descriptive and uses the _data convention that we've established for instances.
No functionality in the module has changed.
* Add oVirt module to manage VMs
This patch add oVirt module to manage Virtual Machines
* Add oVirt module to manage authentication
This patch add oVirt module to manage authentication
* Add oVirt module to manage disks
* Added VM state management and fixups
* Modification of describe_gateways key so that it is consistent with what create_gateway returns.
Also added AnsibleModule spec to require bgp_ip on state=present as defined in the doc
* Don't remove CustomerGateways key to preserve backward compatibility
The 'short_description' in netapp_e_lun_mapping was a
list instead of txt.
This fixes errors on 'ansible-doc -l' of form:
ERROR! module netapp_e_lun_mapping has a documentation
error formatting or is missing documentation
Fixes: #17634 (ansible/ansible)
When using `use_max` or `use_min` in `pam_limits`, the new value is an integer compared with the actual_value which is a string, so they are always different and the module reports a changed but none occurred.
The modules prefixed with netapp_e* are built to support the
SANtricity storage platform.
The modules provide idempotent provisioning for volume groups, disk
pools, standard volumes, thin volumes, LUN mapping, hosts, host groups
(clusters), volume snapshots, consistency groups, and asynchronous
mirroring.
They require the SANtricity WebServices Proxy.
The WebServices Proxy is free software available at
the NetApp Software Download site:
http://mysupport.netapp.com/NOW/download/software/eseries_webservices/1.40.X000.0009/
Starting with the E2800 platform (11.30 OS), the modules will work
directly with the storage array. Starting with this platform, REST API
requests are handled directly on the box. This array can still be
managed by proxy for large scale deployments.
* letsencrypt: improve error handling
Use the new "body" field of the info dict in case of a HTTPError.
* letsencrypt: HTTP 202 is a valid status while polling
The domains method was not defined, and therefore when specifying
a parent domain during route domain creation, the process would
fail.
Tests have been added to detect this going forward
* Fix bug in processing of null return
* Fix multi-dc folder location by enhancing the foldermap and using it to search
* Remove unused functions
* Refactor finding vm by folder
Fixes#2900
* Use common file arguments on destination file
* Rename 'compression' to 'format' h/t @abadger
* Add support for plain 'tar' format
* Ensure check_mode is respected
* clean up functions and remove unneeded code
* config difference now includes keyword argument
* module reports changed when save argument is yes with or without check_mode
* updated fail_json return with exc kwargs
* fixed up import statements
* fixes issues with import error
* removes need for filter attribute in Cli instance
* now filters config either from device or provided via config argument
ref: #2890
Jenkins stores the information about the state (disabled/enabled) in the config, which result in a race condition between `config` and `enabled` and we loose idempotency. It makes sense to define them mutually exclusive.
Renamed `enable` to `enabled`. Ansible uses the name `enabled` in many modules, e.g. service as it indicates a state not an action.
Ensure the HAVE_FIREWALLD checks check only for the
presence of the python dependencies, and not the age
of the library or the state of the service, which
are checked later.
This bug was introduced accidentally when refactoring to 2.2. The instance
of the candidate config was deleted. This adds the candidate config
instance back
fixes#2890
current implementation was breaking making the module unusable, changing to the list comprehension fixes this. Also default to seconds instead of throwing a exception when no duration units are supplied as this causes tests to fail
Adds bigip_ssl_certificate module
This module is another in the ongoing "bootstrapping saga" that is
being undertaken. With this module you can manage the lifecycle of
the SSL certificates on a BIG-IP. This includes those used for
SSL offloading.
Tests for this module can be found here
https://github.com/F5Networks/f5-ansible/blob/master/roles/__bigip_ssl_certificate/tasks/main.yaml
Platforms this was tested on are
12.0.0
12.1.0
* removes get_module() factory method for NetworkModule
* add src argument to provide path to config file
* add new choice to match used to ignore current running config
* add update argument with choices merge or check
* add backup argument to backup current running config to control host
* add defaults argument to control collection of config with or withoutdefaults
* add save argument to save current running config to startup config
* commands argument now accepts a dict arguments
* only show commands are allowd when check mode is specified
* config mode is no longer allowed in the command stack
* add argument match with valid values any, all
* zypper_repository add auto_import_keys options
* also give more output on failure (rc, stdout, stderr)
* be more specific in the doc for auto_import_keys
* add runrefresh option to zypper_repository
* this comes out of ansible/ansible-modules-extras#2411, where AnderEnder adds refresh to the zypper module
* adds a way to force zypper to refresh a repository
* can be used to refresh independently of auto_import_keys
* add option to run name=* runrefresh=yes
* name runrefresh to not break existing use to refresh (now alias to autorefresh)
* add version_added flag to autorefresh
* remove wrong version_added comment
Recently, a user reported that the bigip_facts module was failing with the error
received exception: object of type 'itertools.imap' has no len()
This reported was occurring at line 1657 of the bigip_facts module
bug report is here
https://github.com/F5Networks/f5-ansible/issues/25
Upon further investigation, the map function for returning the specified
includes was returning an iterator, and calling len() on an iterator does
not work.
I believe this problem was caused by part of the Python 3.x effort insofar
as the inclusion of this line
https://github.com/ansible/ansible/blob/devel/lib/ansible/module_utils/basic.py#L143
seems to affect our usage of map(), probably for the better anyway, and we need
to change our expectations in our module's code to no longer assume a list, but
instead assume an iterator.
After trawling through the module_utils/basic code, I think a list
comprehension is more appropriate here anyway, so I'm changing it to be
that. The affected user reported it works this way, and my own testing
on 2.2.0 supports that.
* Add facts and storage system modules.
* Update version metadata.
* Add facts and storage system modules.
Update version metadata.
Add init files.
* Fixes for review comments.
* Fixes for review comments.
* Update document fragments for api_* options to indicate they are required.
* Correct bad string concatenation.
* Add option description for validate_certs since basic_auth arg_spec is no longer extended.
* Add default value for validate_certs in docstring.
* Rename directory name so it applies to netapp and not simply eseries platform.
* Rename modules to differentiate other netapp modules.
