When creating a new account, check to see if the expiration parameter is negative and pass in the appropriate parameter. Since the negative integer passed into expires is converted to time.struct_time which in turn gets converted to a formatted time string when passed to the underlying command, a -1 or large negative number would result in passing a date before 1970-01-01 to the underlying command.
This had the opposite effect of creating an account with no expiration account resulting in a newly created account that was already expired, or just throwing an error on certain systems.
* gather_subset is a list of strings
When gather_subset is an integer, a message pointing out the problem, current tb error
is replaced by this one:
ERROR! the field 'gather_subset' should be a list of (<class 'str'>,), but the item '42' is a <class 'int'>
...
* gathering_facts test: ensure smart gathering is on
* Currently network_cli support multiple prompts
single answer as response. This PR adds support
for multiple answers.
* In case of multiple prompts and mulitple answers the
index of a particular prompt in the prompts list should
match with the index in the answer list.
In python 3.7.0, changes in `ssl.py` breaks `smtplib.SMTP_SSL`, which
then breaks `mail` module in ansible.
Run this line in python shell:
import smtplib;smtplib.SMTP_SSL().connect(host='smtp.gmail.com', port=465)
Before python 3.7.0, we will get:
(220, b'smtp.gmail.com ESMTP j13-v6sm3086685pgq.56 - gsmtp')
In python 3.7.0, we get such error at `lib/python3.7/ssl.py` line 843, method `_create`:
ValueError: server_hostname cannot be an empty string or start with a leading dot.
The ssl module is using host info on SMTP_SSL instance, which is not set.
The fix/workaround is simple, just pass host info to it:
import smtplib;smtplib.SMTP_SSL(host='smtp.gmail.com').connect(host='smtp.gmail.com', port=465)
Fixes: #44550
Signed-off-by: Guo Qiao <guoqiao@gmail.com>
This prevents the accidental creation of TXT records where every
single word gets split into its own string, such as TXT record values
in the format of `"foo" "bar" "baz"`. That being an implicit behavior
I have very hard to see anyone purposely relying on.
TXT record values can still explicitly be defined as one or more
strings, without any change in syntax.
Resolves#43380
* adding next network pluggin
* include exclude option for next_ip search
* changing example provider input
* adding new line at end
* version added info
* to fix shippable errors
* to fix shippable errors
* adding exclude option in doc section
* fix review comment
* fix review comment
* Correct the default doc for attached in ec2_eni
Also corrected a typo in the summary
* Address ansible-test sanity error about E324
* Fix and remove the E325 suppression for ec2_eni
Extends `module_defaults` by adding a prefix to defaults `group/` which denotes a builtin list of modules. Initial groups are: `group/aws`, `group/azure`, and `group/gcp`
* add route module
* add test
* add table
* add route table
* fix dict
* fix
* fix
* route table accept no name
* add default
* fix
* fix
* fix
* fix pep
* support route table
* fix line ending
* fix pep
* fix
* fix
* set the default to 'None'
* make return value only id
* camel and snake
* set test alias
* change facts line ending
* change test
* fix
* add default
* fix
* fix line break
* remove unsafe args
* fix comment
* fix
* adding postgresql server facts
* updating postgresql server facts, minor changes
* changed return dict to list
* fixed test
* fixed several issues
* several updates
* fixed naming
* list -> complex
* try again
* fixed mistake
* added user_visible_state to the result
* added fully_qualified_domain_name
* fixed second test
* updates
* fixed test
* updated samples in return value
* wait for postgresql server to be actually ready
* another approach for test
* second server was not created
* fixing pr comments
* fix sanity
* removed forgotten ansible_facts
* added account_alias in the response of module aws_caller_facts
* added comment to explain list_account_aliases
* renamed caller_identity to caller_facts as the content is extended
* created changelog
* security-policy needs the iam:ListAccountAliases for this module to work
* test now checks for the added field account_alias
* gracefully handle missing iam:ListAccountAliases permission
* Added helper module for generating ACME challenge certificates.
* Soft-fail on missing cryptography. Also check version.
* Adding integration test.
* Move acme_challenge_cert_helper from web_infrastructure to crypto/acme.
* Adjusting to draft-05.
* The cryptography branch has already been merged.
Wow, this does not seem to be an uncommon misspelling. Might be there
are some left that span over two lines. I noticed the one in the git
module and then used `grep -rw 'the the'` to find some more.
Many OpenStack modules provide the ability to register a resource in a
project other than the one being used to authenticate with, by adding a
project parameter to the module. Examples include os_network, os_subnet,
and os_router. This change adds a project parameter to the
os_security_group module.
Fixes: #34467Fixes: #30292
This allows to parse the output when the user's locale changes the
commands' output. For example chkconfig uses 'Ein' and 'Aus' instead of
'on' and 'off' when using LANG=de_DE.UTF-8 breaking the service
detection on RHEL 6.
* Docker: Convert units for kernel_memory
Other memory arguments already use human_to_bytes to convert the
units. This change makes the behavior more consistent.
* Update Docker unit parsing documentation
This explains the units more precisely and makes the documentation
more consistent.
This fixes#16526.
when using only an activation key without any channels.
As already suggested by mattclay in
https://github.com/ansible/ansible/pull/25079
and also patch unit test for rhn_register and
add test case for activationkey only
Add _flushed_hosts dict to store when handlers are flushed and prevent them
to be executed too early using _filter_notified_hosts.
Add _wait_on_handler_results to wait only for handlers to be completed.
Remove only hosts that have been flushed from handler notified list.
Fix#31504, #23970
Fixes#43653
- Change descrtiption parameter default to null.
When reading an empty description from AD, powershell returns a null.
- Convert after and before diff keys to ordered hashes.
Unordered state descriptions make --diff report equal values as
changes.
OpenNebula 5.5.8 expects int in DISK_SIZE field when you create a vm, not a float.
For example, "42 GB" should be sent to OpenNebula API as "43088", not as "43088.0". MEMORY you cast to int explicitly, but DISK_SIZE not.
Portions of the f5-sdk were removed as well as the netaddr library
and were replaced with libraries that are part of ansible. Additionally,
deprecated code has been removed.
Cloudflare's DNS API deals with the case insensitivy of DNS names and
IPv6 addresses by forcing them into lower case. To properly be able to
detect changes the Ansible module needs to behave accordingly.
To what extent the API expects sent DNS names to be lower case varies
between record types. Yet, since sending lower case always works, and
since we always get lower case back, it feels cleanest to use lower
case for all DNS names.
* Use newer is_sequence function instead of MutableSequence. Fixes#44327
* Add changelog for #44331
* Update changelog fragment to describe the fix in more detail
Fixes#40650Fixes#40245Fixes#41541
* Refactor netconf_config module as per proposal #104
* Update netconf_config module metadata to core network supported
* Refactor local connection to use persistent connection framework
for backward compatibility
* Update netconf connection plugin configuration varaibles (Fixes#40245)
* Add support for optional lock feature to Fixes#41541
* Add integration test for netconf_config module
* Documentation update
* Move deprecated options in netconf_config module
* init remote_vv_file
* update documentation
* update documentation 2
* update description
* add correct style of ovirt info
* update documentation
* correct line width in docs
* use correct end line in docs
* Add WTI OOB and PDU Device status, control and configuration module
* reorganized module names, removed multiple modules
* Fixed sanity check failures
* Fixed sanity check failures
* Fixed sanity check failures
* Fixed sanity check failures
* moved module from network directory to remote_management.
added validate_certs parameter
added example returned data.
* pep8 issues
* cpm_status code removed from module section and placed in plugins/lookup.
Modified module code to lookup style code
* EXAMPLE spacing format error
* EXAMPLE formatting
* removed split_line parameter. Return is exclusively formatted JSON
* display edits
* delete display debug text
* Change the examples to use YAML rather than k=v shorthand
* re-enabled "required" flags for cpm_username and cpm_password
* moved redundant ansible.module_utils._text import of to_native to previous ansible.module_utils._text import line
* Got unusual error from GitHub checkin of Unstable, bump to see if it happends again.
* New boto3_facts module
boto3_facts aims to help users see whether their python and module
versions are as expected.
* Rename to `assert_python_requirements`
* Update integration tests
* Document options
* fix imports
* boilerplate
* fix docs
* reorder import
* Make distutils optional and fail gracefully when it is not available
* fix example doc
* fix docs on requirements_facts
* Don't use copy.deepcopy in high workload areas, use deepishcopy. ci_complete
* Add tests
* Add changelog fragment
* rename to naive_deepcopy and add extra docs
* Rename to module_response_deepcopy and move to vars/clean
One can install alternate packages managers on debuntu machines.
However, doing so doesn't mean you want to suddenly start using them.
Add in a check similar to the fedora yum/dnf check that sets apt as the
pkg_mgr if the ansible_os_family is Debian.
This commit introduces a new module called vultr_network_facts.
This module aims to return the list of networks avaiable in Vultr.
Sample available here:
```
"vultr_network_facts": [
{
"date_created": "2018-08-02 11:18:49",
"id": "net5b62e8991adfg",
"name": "mynet",
"region": "Amsterdam",
"v4_subnet": "192.168.42.0",
"v4_subnet_mask": 24
}
]
```
The documentation defines `cpu` as:
```
default cpu type
default value: kvm64
```
Therefore defining `cpu` with `8` in the `examples` section makes no
sense. So changing to `cores`.
Signed-off-by: Christian Rebischke <Chris.Rebischke@posteo.de>
When a datacenter has multiple clusters it is required to use the
cluster name to find the correct default resource pool to import
OVF images to.
Fixes issue 42644.
Signed-off-by: Michael Cronenworth <mike@cchtml.com>
* add example using the "when" keyword
<!--- Your description here -->
add example using the "when" keyword comparing group_var variable to returned value from ios_facts
+label: docsite_pr
+label: issue #311
* Update ios_config.py
* Update ios_config.py
* Refactor yum and dnf, add feature parity
Signed-off-by: Adam Miller <admiller@redhat.com>
* remove unnecessary module_utils, move the classes into the module code
Signed-off-by: Adam Miller <admiller@redhat.com>
* remove yum -> yum4, out of scope
Signed-off-by: Adam Miller <admiller@redhat.com>
* use ABCMeta
Signed-off-by: Adam Miller <admiller@redhat.com>
* re-arrange run() caller vs callee
Signed-off-by: Adam Miller <admiller@redhat.com>
* make sanity checks happy
Signed-off-by: Adam Miller <admiller@redhat.com>
* fix yum unit tests
Signed-off-by: Adam Miller <admiller@redhat.com>
* remove unecessary debug statements, fix typo
Signed-off-by: Adam Miller <admiller@redhat.com>
* fix licensing and attribution in yumdnf module_util
Signed-off-by: Adam Miller <admiller@redhat.com>
* include fix from PR 40737
original commit 5cbda9658a
original Author: Strahinja Kustudic <kustodian@gmail.com>
yum will fail on 'No space left on device', fixes#32791 (#40737)
During the installing of packages if yum runs out of free disk space,
some post install scripts could fail (like e.g. when the kernel
package generates initramfs), but yum would still exit with a status
0. This is bad, especially for the kernel package, because it makes
it unable to boot. Because the yum module is usually used for
automation, which means the users cannot read every message yum
prints, it's better that the yum module fails if it detects that
there is no free space on the disk.
Signed-off-by: Adam Miller <admiller@redhat.com>
* Revert "fix licensing and attribution in yumdnf module_util"
This reverts commit 59e11de5a2.
* move fetch_rpm_from_url out of yumdnf module_util
Signed-off-by: Adam Miller <admiller@redhat.com>
* fix the move of fetch_rpm_from_url
Signed-off-by: Adam Miller <admiller@redhat.com>
Make git module support --separate-git-dir option. When git version is higher than 1.7.5, use built-in --separate-git-dir option during the clone. When lower, adjust the location of git dir manually after clone to achieve the same effect.
The inventory plugin api grew a self.cache that wasn't there when we
first wrote it. There's also a need to pull in the documentation
fragments so that we have the cache parameters.
Due to blank line returned from vCenter, wrong values were appended
to NTP server list. This fix adds filter for such blank values of NTP servers.
Fixes: #44183
Signed-off-by: Abhijeet Kasurde <akasurde@redhat.com>
Grammar on the description for option 'list' was grammatically incorrect. Also suggest to add wording that clarifies that 'list' is indeed Access-Based Enumeration.
+label: docsite_pr
* Add note about adding new key requiring update
When a new repo and new key are added, attempts to install packages
signed by that key fail until `apt-get update` is run. This note
is an attempt to help users avoid getting errors when they miss
this step.
* related to issue #25091
* switch example to apt module
* Compress src interface into one key
* Modified regex to support varied interface types
* Fixed documentation
* Unpacking return values from splitting method
- support config operations for EXOS-based platforms
- add regex to detect command failure responses
- add exos action plugin for "backup" operation
- add unit tests for exos_command (currently 94% coverage of
exos_config.py)
* Add backup option
* Only backup shadow file when the OS has one
* Only backup shadow file for SunOS
* Update docs on backup feature
* Add changelog fragment
* Add tests for shadow backup
* Remove backup option, make it automatic
Remove the option to enable/disable backups and make it automatic. Add note to docs describing this behavior.
Change tests to account for new module behavior.
Change section name in changelog fragment since minor_features is not a valid section.
NOTE:
1. use os.open() with os.O_CREAT|os.O_EXCL to check existence
and create a lock file if not exists, it's an atomic operation
2. the fastest process will create the lock file and others will
wait until the lock file is removed
3. after the writer finished writing to the password file, all the reading
operations use built-in open so processes can read the file parallel
* strip additional comments from /etc/default/passwd
Strip trailling comments from /etc/default/passwd like
MINWEEKS=1 #MINWEEKS=2
MAXWEEKS=12 # MAXWEEKS=8
Which otherwise cause failures with "failed to read /etc/default/passwd: too many values to unpack"
* fix carriage return typo in commit
* yet another typo in commit
* Fix indent problem
* add changelog fragment for PR 43931
Ideally I would have liked to compare the TTL as part of the
prerequisite check. Sadly that isn't supported by the RFC 2136 update
protocol. Hence the additional query.
Resolves#39465
* Update cloudflare_dns account link
* Add SSHFP and TLSA records to cloudflare_dns module
These are record types which Cloudflare recently added support
for. They both go well together with DNSSEC.
Technically it's a bit of a simplification to use the hash_type
parameter for TLSA records. Yet, it fits with all the real world usage
I have seen, and it keeps the module from sprawling too much.
Related to #43803
* Update netconf plugins for junos and iosxr
Fixes#39160
* Update api signature for netcon api's
from variable arguments to named arguments
* Udpate get default rpc supported
* Fix CI issue
* restore task arg splatting
* reverts #41804
* supersedes #41295
* fixes#42192
* after lots of discussion amongst the core team, we decided to preserve this feature, clarify the runtime warnings/docs, and prioritize a path toward fixing the underlying behavior that causes this feature to be insecure (un-namespaced facts).
* update faq text
note that warning is disabled when inject_facts_as_vars is
* wordsmithing FAQ entry
As of today, self.returns it not the source of truth. If the return
value from querying the resource contains more values than the one
listed in self.returns, those value will be returned even though not
explicitly specified in self.returns.
This commit ensures that only the values listed on self.returns are
actually returned. The other values not listed are supressed.
Every scaleway modules will require the same base of parameters. Instead
of documenting it each and every time, we build a doc fragment that will
allow one to simply extend documentation for every new module.
extends_documentation_fragment: scaleway
During the installing of packages if yum runs out of free disk space,
some post install scripts could fail (like e.g. when the kernel package
generates initramfs), but yum would still exit with a status 0. This is
bad, especially for the kernel package, because it makes it unable to
boot. Because the yum module is usually used for automation, which
means the users cannot read every message yum prints, it's better that
the yum module fails if it detects that there is no free space on the
disk.
* Common Rest Module using httpapi transport
* comparing json objects
* platform independant Rest Api
* FTD rest module
* Remove FTD module which will be committed later from cisco team
* remove debug code
* remove common rest module
* fix shippable errors
* Add notification users
* more shipable errors
* 29891 use os.path.realpath to follow keyfile symlinks
* 29891 add parameter follow
* updated changelog fragment
* add documentation and set default to false
* pip: combine chdir and env only when env is set
This fixes an AttributeError when chdir without virtualenv is specified:
File "/tmp/ansible_2UAFsZ/ansible_module_pip.py", line 387, in main
env = os.path.join(chdir, env)
File "/usr/lib64/python2.7/posixpath.py", line 75, in join
if b.startswith('/'):
AttributeError: 'NoneType' object has no attribute 'startswith'
* Add test for pip with chdir
Signed-off-by: Till Maas <opensource@till.name>
* Only template values in vars_prompt rather than all vars
This allows the use of variables in vars_prompt fields but allows variables entered in the prompt to affect play vars rather than throwing an undefined error.
Only post validate if there was a vars_prompt
* Add tests for vars_prompt
* Check the password format
Check the password format and notify user if they
input unencrypted password.
* Fix sanity error
* Add integration test
* Missed a task name
* Hard code the testing password
Since some testing platfrom has no passlib installed
* Add changelog fragment
* Rework some English sentences
* Fix a grammar mistake
The Vultr API is inconsistent in the type of the value it returns
based on the resources. While most of the time it will be a dict, for
some resources it will be a list (/v1/user/list, /v1/block/list).
query_resource_by_key() fails if the return value isn't a dict (.items()
does not exist on list). This patch aims to support both list and dict.
* Hold httpapi response in BytesIO
* Let httpapi plugin deal with HTTP codes if it wants
* Python 3.5 won't json.loads() bytes
* Don't modify headers passed to send
* Move code handling back to send()
but let httpapi plugin have a say on how it happens
The bundled selectors library which is used by the local and ssh
connection plugins had a bug which caused a traceback in a cornercase.
If selectors were in use and a syscall was interrupted, selectors would
attempt to restart the syscall after the interrupt was processed. if
the attempt determined that the timeout for running the syscall had
already expired, the code attempted to raise OSError. The raise was
using a Python3-ism and needed to be ported to work on Python2.
Fixes#41630
* Fix cli_command junos test failure and update doc
* Fix cli_command module integration test failure
for junos
* Update cli_command module doc for prompt and
config command run scenario's
* Update cli_command module doc
* Collecting PEM -> DER conversions.
* Using cryptography instead of OpenSSL binary in some situations.
* Moving key-to-disk writing for key content to parse_account_key.
* Rename parse_account_key -> parse_key.
* Move OpenSSL specific code for key parsing and request signing into global functions.
* Also using cryptography for key parsing and request signing.
* Remove assert statements.
* Fixing handling of key contents for cryptography code path.
* Allow to disable the use of cryptography.
* Updating documentation.
* 1.5 seems to work as well (earlier versions don't have EC sign function). Making Python 2.x adjustments.
* Changing option to select_crypto_backend.
* Python 2.6 compatibility.
* Trying to test both backends separately for acme_account.
* Also testing both backends separately for acme_certificate and acme_certificate_revoke.
* Adding changelog entry which informs about select_crypto_backend option in case autodetect fails.
* Fixing YAML.
Fixes#42310
Previously, the firewalld module was making a call to
FirewallClientConfig.getZoneNames() which doesn't exist in versions
of firwalld older than 0.4.2, this patch implements the same logic
with older API calls to not require a newer version of firewalld.
Signed-off-by: Adam Miller <admiller@redhat.com>
In query_resource_by_key(), there is an equal comparison that is made to
know if the object we are looking for is present. Due to type difference
this comparison doesn't always retrieve true, even when it should.
This is due to the fact that the value in r_data dict are of type
unicode, while the other can be of type int, float,... .
```
>>> a = u'1'
>>> type(a)
<type 'unicode'>
>>> b = 1
>>> type(b)
<type 'int'>
>>> a == b
False
>>> str(a) == str(b)
True
```
Hence the values, for comparison purposes, are casted into strings.
* Merge again trickily similar Accelerated networking and IP forwarding
* Add type to enable_ip_forwarding documentation
* Fix merge error
* auth to auto
* azure_rm_networkinterface: remove auth_source from tests
* azure_rm_networkinterface: remove spurious auth_source from test
* azure_rm_networkinterface: Revert formatting on test
* azure_rm_networkinterface: Correct indentation
_parsed_return_data should only be used with the return from a module.
This location was invoking a remote shell command rather than a module
so we don't want it here.
they dropped the Linux so now it only shows as ALT, it should still be backwards compatible
pkg_mgr detection relies on `Altlinux` string, so properly setting os_distribution should take care of it as side effect
fixes#43539
* Remove use of simplejson throughout code base. Fixes#42761
* Address failing tests
* Remove simplejson from contrib and other outlying files
* Add changelog fragment for simplejson removal
* Change how data is sent to the persistent connection socket.
We can't rely on readline(), so send the size of the data first. We can
then read that many bytes from the stream on the recieving end.
* Set pty to noncanonical mode before sending
* Now that we send data length, we don't need a sentinel anymore
* Copy socket changes to persistent, too
* Use os.write instead of fdopen()ing and using that.
* Follow pickle with sha1sum of pickle
* Swap order of vars and init being passed to ansible-connection
