* Add check_mode to get_url that does a HEAD request to make sure the URL exists, but doesn't write the real file
* Add info about new --check behavior to docs. Add tests for the new behavior. Populate res_args with the info the tests are looking for.
* Add trailing comma
* Change nonexistent test URL to http://{{httpbin_host}/DOESNOTEXIST. Fix spacing while I'm at it
* Further spacing cleanup
* State that this functionality is in Ansible 2.4+
* Big testing doc refactor
* Combine all the testing documentation in to one place to make it easier to find
* Convert everything to RST
* Create testing_network guide
* Create testing landing page
* For each section detail "how to run" and "how to extend testing"
* More examples
* Lots more detail
/usr/sbin/locale-gen differs between Debian and Ubuntu. With Debian,
locales must be added to /etc/locale.gen before calling locale-gen
command. With Ubuntu, /etc/locale.gen is updated by
/usr/sbin/locale-gen.
Use the locale-gen module which handles both distributions.
* PostgreSQL tests: use package module
* PostgreSQL tests: use ansible_service_mgr in order to test init system
* PostgreSQL tests: fix a typo
* PostgreSQL tests: install only required packages
* PostgreSQL tests: use role default value when unchanged
* Run `save` before exiting config mode.
* Fix unit tests for `save`
* Allow `save` to be on its own again and introspect success
* Introspecting `compare running` makes this a lot harder.
Move `save` tests to integration tests
Added an integration test for fetch module idempotence. (Testing
that validate_checksum is doing what it's supposed to is harder as we'd
have to create a race condition with the downloaded data to trigger it.
Probably need to make that a unittest eventually).
Also give a deprecation message to the validate_md5 parameter so that we
can eventually get rid of it.
* keep unsafe .. unsafe
fixes#23734, which was broken in previous fix that allowed non string types to be templated
use new 'is_template' function vs bastardizing others
refactored clean_data to allow for arbitrary data structures to clean
fixed/removed some tests
* deal with complex data for is_template
* typos
* Fix vault reading from stdin (avoid realpath() on non-links)
os.path.realpath() is used to find the target of file paths that
are symlinks so vault operations happen directly on the target.
However, in addition to resolving symlinks, realpath() also returns
a full path. when reading from stdin, vault cli uses '-' as a special
file path so VaultEditor() will replace with stdin.
realpath() was expanding '-' with the CWD to something like
'/home/user/playbooks/-' causing errors like:
ERROR! [Errno 2] No such file or directory: u'/home/user/ansible/-'
Fix is to specialcase '-' to not use realpath()
Fixes#23567
* to_text decrypt output when writing to stdout
* Build HTTPSClientAuthHandler more similarly to how HTTPSHandler works
* Add docs for new client cert authentication
* Support older versions of python
* Simplify logic
* Initial support for client certs in urls.py
* Add an extra test
* Add a get_url test for client cert auth
* Add additional test for client cert auth, with validation and ssl mismatch
* Skip assert when http tester not available
* Update version_added for new options
Copy module was walking over files in subdirectories repeatedly (a
directory tree a few levels deep could bring the time spent into the
tens of minutes)
This was traced to the fix for this bug report: https://github.com/ansible/ansible/issues/13013Fixed#13013 a different way and added an integration test to check for
regressions of #13013 as we optimize this code.
Fixes#21513
* Check for proper response key on eos_banner map_config_to_obj
If we run the task with 'login' banner, the 'show banner' command
will return a dict containing key 'loginBanner'.
However for motd, it will just return 'motd'.
Yay naming consistency!
* Do not assert session exists on eos_banner response not changing device
* Add workaround for evasive in apache2_module
* Fixes#22635
* Clean up workarounds for php/shib
* Add test for evasive workaround
* Remove use of re module, since all searches work with native python
* Add unit tests to apache2_module name replacements
Go back to using re package where needed
* Rename replace_name to create_apache_identifier
Test that, when a group is created with group_by, variables from the
corresponding group_vars/ file are correctly pulled in, and override
variables specified in group_vars/all.
See ansible/ansible#8664.
* fix 'sequence' lookup shortcut syntax and documentation
* Update playbooks_loops.rst
Minor edits for grammar and clarity.
* Update playbooks_loops.rst
Another tweak for clarity.
* Fixes bug https://github.com/ansible/ansible/issues/22455 - Git reset fails when file with the name 'origin' is present in the repository
* #22455 Integration test for git module to test if it clones a repository with a file named origin
* Fixed yaml errors for test/integration/roles/test_git_clone/tasks/main.yml
* Removed the newly added target git and moved the code to test/integration/targets/git
* Reorganised code structure as per comments
* Fixed playbook name
* Added code that didn't make it to the playbook before.
* No longer need git.yml
* Remove non tasks specific code from tasks file and use existing variables
* Removed msg var from the vars.yml
* update integration tests for updated boto exception message
* integration tests fail on both "test credential" test cases
exception bubbles out of module. instead catch and wrap
* ec2_group does not support updating a security group's description
AWS security group descriptions are immutable.
if ec2_group finds a group that matches by name, but the descriptions do not match, the module does not support this case
previously it would check if the group was used, but would not do anything if it was
old behavior was erroneous because it could make a user expect that the description change of a group was fine when in fact it did not occur
also, it made an expensive check against all ec2 instances for no good reason
* comments not doc strings
* else must have pass w/o doc-string statement
* Catch specific BotoServerException, give context around error when fetching SGs
* python3 compatible exception blocks
* add traceback to fail_json
* two blank lines before first function
* fixes#15770
* When running under the UTF-8 codepage, Powershell subprocesses will fail (eg, Start-Job, others) if the input encoding is using the default BOM preamble. This fix forces it to use no preamble in leaf_exec and win_shell, and includes tests to verify that Start-Job works.
This change to the template action plugin make template use the
platform's native newline_sequence for Jinja.
We also added the option `newline_sequence` to change the newline
sequence using by Jinja if you need to use another newline sequence than
the platform default.
This was previously discussed in
https://github.com/ansible/ansible/issues/16255#issuecomment-278289414
And also relates to issue #21128
* fixes#22441
* fixes#22655
* moves all env handling into the exec wrapper; this should work for everything but raw, which is consistent with non-Windows.
* win_scheduled_task: Added frequency: once and check_mode support
This patch includes:
- Renamed `execute:` parameter to `executable:`
- Renamed `argument:` parameter to `arguments:`
- Implemented `frequency: once` support
- Implemented check_mode support
- Fix idempotency issue related to empty description
- Added integration tests
* Improve the integration test structure
I think this is a great way to test normal mode and check-mode from the same playbook.
* Small fixes after review
Fixes#22575 - issue under new exec wrapper where unconstrained handle inheritance (for stdin) caused WinRM to block on breakaway processes. Uses explicit handle inheritance to ensure that only stdin read handle gets inherited. Adds test to ensure that async is actually async.
* candidate var can now be string, list or element
* fixes minor bug in junos_template for backup argument
* disabled invalid integration test for junos_template
* removes cli functions
* adds comment and confirm to arguments
* implements zeroize argument
* fixes get_diff function in junos shared lib to return diff
* lots of minor bug fixes in junos_config
* minor syntax fixes in junos_netconf
* updates netconf integration tests
* eos_banner docs improvements
* run eos_banner tests
* basic eos_banner tests
* remove trailing newlines
* remove hack
* Various eos_banner fixes
No unicode tests, as the device doesn't support it
Correctly set provider
result.command should be empty list if no changes
* test eos_banner over eapi
* revert docs changes, moved to separate PR
* description is only required when group state is present
also note that AWS requires a non-empty description when creating a security group
* clarify description requirement
fixes integration test to check for full rpc-reply string when
converted from xml. This is a change from 2.2 as the rpc-reply tag
is no longer stripped off
* Fixes#21316, add testcase based on this
* Add option `--branch NAME` to git clone command in case of branch or
tag in combination with depth=1
* This option should work back to at least git 1.8 and thus on all
supported distributions
* Provide better warning if depth is dropped
* VLAG module and tests
* Correct the link in the description of the module
* Move roles to integration/roles directory and remove aliases file.
* update module short description
* Change module directory name to Lenovo
* change import cnos statement
* Initial commit of cnos_conditional_template
* fix compile error in module, change module short description, move roles to integration/roles
* fix line length
* Change module directory name to Lenovo
* change import cnos statement
* Initial commit of cnos_condition_command
* fix pep8 error of line too long
* fix pep8 tab/space issue
* Move roles to integration/roles directory and remove aliases file.
* Update module short description
* Change module directory name to Lenovo
* change import cnos statement
* initial comment of cnos_save
* Move roles to integration/roles directory and remove aliases file.
* update the module short description
* Change module directory name to Lenovo
* change import cnos statement
* fix in cnos.py utility file
* fix compile errors in cnos.py, update module short description in cnos_vlan.py, move vlan roles under integration/roles
* corrections in readme
* Change module directory name to Lenovo
* change import cnos statement
* initial commit
* fix tab/space issue
* fix space issue in vars/main.yml
* Move roles to integration/roles directory and remove aliases file.
* update module short description
* change module directory name to Lenovo and change import cnos statement
* cnos_portchannel initial commit
* Fix line length issue in module
* fix whitespace on line 90
* correct yamllint space issue
* Move roles to integration/roles directory and remove aliases file.
* update module short description
* Change module directory name to Lenovo
* change import cnos statement
* WIP: wait_for_connection: Wait for system to be reachable
This action plugin allows to check when a system is back online and
usable by Ansible.
As an example, when doing a SysPrep and running Enable-WinRM.ps1, it
takes between 10 to 20 seconds between the WinRM TCP port to open, and
it actually being able to server Ansible requests. This time is variable
and depends on the boot process.
Current implementation is specific for Windows (WinRM) only, this will
be fixed shortly.
This fixes#19998
* Support other transport types
* Various improvements
- Fix reported typo
- Add transport_test support in accelerate plugin
- Ensure port is an integer
* Improve examples
* Small fixes
- Use correct ConfigureRemotingForAnsible.ps1 script name
- Only use win_ping when remote shell is known to be Powershell
- Add integration tests to CI framework
* Make sure the uri module warns about using the deprecated HEADER_ arguments, so that we can actually deprecate in a future version
* Use module.deprecate
* No need to de-dupe ourselves, looks like it is done somewhere I didn't initially see
* Adds integration test steps to win_regedit test to cover Extras 2090
* Fixed merge conflict and error in win_regedit integration tests
* fixed pep8 space after comma requirement discovered by ci build
* Fix vyos signatures to match new versions
* Fix test cases referring to 'updates' instead of 'commands'
* I think this is an artifact of `connection: network_cli`?
* module will now check protocol running state before return
* warns on config argument no longer being needed
* lots of integration test cases updated
* updates unit and integration test cases for state check
fixes#21908
eos_config module wasn't respecting config block path (parents). This
patch fixes that problem. Also fixes a number of integration tests
cases
fixes#21903
Fix 'task name is not templated in retry callback'
Add a task_name property to TaskResult that knows to
check in TaskResult._task_fields.
Add integration test for v2_retry_runner callback
Fixes#18236
When parsing the configuration, the wrong indent level was set so the
configurations could not be properly compared. Also noted that
defaults: yes needed to be added to the integration tests to make the
test cases valid, which is also included in this patch
fixes#21828
* First set of CNOS network modules
Two modules: cnos_vlan and cnos_command. First 2 of 17 CNOS modules
* Revert "First set of CNOS network modules"
This reverts commit 63da68b2a857e622b449f514050eed25fbcf0a35.
* cnos_vlan module
First of 17 CNOS modules
* Modify playbook for cnos_vlan module testing
* fix issue found by shippable in cnos.py
* cnos_backup module
* Fix Python3.5 compilation error
* remove cnos_backup module from this PR
* add aliases file so that testing of the cnos_vlan module will be skipped.
* Use empty aliases file
* utility scripts under module_utils changed to BSD license.
* clean up PEP8 issues, add cnos_vlan tests
* remove blank lines from yml files
* remove more blank lines
* remove even more blank lines
* Revert "remove even more blank lines"
This reverts commit bb3998661ce387260340bca39d5f146f5ec54f44.
* re-submit last commit
* fix blank lines in this file under parsing
* fix yamllint errors in doc section of cnos_vlan
* Fix errors reported by Sanity check using validate-modules
* change test playbook to include tasks for all modules
* fix additional pep8 errors
* more fixes, plus fix for python3
* fix for empty options section, removed unnecessary documentation
* fix up documentation section
* Correct options capitalization
* fix example section, add options
* Fix example section of the documentation
I have from time to time a need of random password without
wanting to write them down (one example is mailman list creation,
that requires a password to be given to be sent to the list owner).
But using /dev/null do not return null, but the empty string, which
doesn't generate a password at all and so do not achieve my use case.
hashlib hashes operate on byte strings. When given a text string on
Python3, hashlib backtraces. When given a text string on Python2,
hashlib will backtrace if the string contains non-ascii characters.
Encode the text string to utf-8 prior to hashing to avoid this problem.
Fixes#21452
* known_hosts: support --diff
* known_hosts: support --diff also without --check
* Add unit tests and fix incorrect diff in one corner case
Tests are good!
* Refactor for readability
* Python 3 compat
* More Python 3 compat
* Add an integration test for known_hosts
* Handle ssh-keygen -HF returning non-zero exit code
AFAICT this is a bug in ssh-keygen in some newer OpenSSH versions
(>= 6.4 probably; see commit dd9d5cc670):
when you invoke ssh-keygen with -H and -F <host> options, it always
returns exit code 1. This is because in ssh-keygen.c there's a function
do_known_hosts() which calls
exit (find_host && !ctx.found_key);
at the end, and find_host is 1 (because we passed -F on the command line),
but ctx.found_key is always 0. Why is found_key always 0? Because the
callback passed to hostkeys_foreach(), which is known_hosts_hash(),
never bothers to set found_key to 1.
* This test does not need root
* Avoid ssh-ed25519 keys in sample known_hosts file
Older versions of OpenSSH do not like them and ssh-keygen -HF
aborts with an error when it sees such keys:
line 5 invalid key: example.net...
/root/ansible_testing/known_hosts is not a valid known_hosts file.
* Fix Python 3 errors
Specifically, the default mode of tempfile.NamedTemporaryFile is 'w+b',
which means Python 3 wants us to write bytes objects to it -- but the
keys we have are all unicode strings.
Network devices in the lab have by default an indentation of 1 for
sub-sections, and so does the netcfg.NetworkConfig constructor
indent param.
This will fix reported issue 21055, and make the integration tests
to pass.
Fixes#21055
- centralized skipping
- also fixed module name broken by previous refactor
- let action modules handle async processing
- moved async into base action class's module exec
- action plugins can now run final action as async
- actually skip copy if base skips
- fixed normal for new paths
- ensure internal stat is never async
- default poll to 10 as per docs
- added hint for callback fix on poll
- restructured late tmp, now a pipeline query
- moving action handler to connection as networking does
- fixed network assumption invocation is always passed
- centralized key cleanup, normalized internal var
- _supress_tmpdir_delete now in _ansible_xxx and gets removed from results
- delay internal key removal till after we use em
- nicer tmp removing, using existing methods
- moved cleanup tmp flag to mking tmp func
* added docs for vault and made trigger shorter: !vault
* added single var valuting
* Update playbooks_vault.rst
Edit pass for spelling and grammar. Ship it!
* Update playbooks_vault.rst
Typo fixes.
* Make the module_utils path configurable
* Add a config value to define the path site module_utils files
* Handle module_utils that do not have source as an error
* Make an integration test for module_utils envvar working
* Add documentation for the ANSIBLE_MODULE_UTILS config option/envvar
* Add it to the sample ansible.cfg
* Add it to intro_configuration.
* Also modify intro_configuration to place envvars on equal footing with
the config options (will need to document the envvar names in the
future)
* Also add the ANSIBLE_LIBRARY use case from
https://github.com/ansible/ansible/issues/15432 so we can close out
that bug.
* win_shortcut: Add missing $check_mode definition
For some reason this entry was missing, possible a merge-conflict gone
wrong :-(
* Added integration tests and bugfix
Add missing changes.
This version just gets the relevant paths from PluginLoader and then
uses the existing imp.find_plugin() calls in the AnsiballZ code to load
the proper module_utils.
Modify PluginLoader to optionally omit subdirectories (module_utils
needs to operate on top level dirs, not on subdirs because it has
a hierarchical namespace whereas all other plugins use a flat
namespace).
Rename snippet* variables to module_utils*
Add a small number of unittests for recursive_finder
Add a larger number of integration tests to demonstrate that
module_utils is working.
Whitelist module-style shebang in test target library dirs
Prefix module_data variable with b_ to be clear that it holds bytes data
* Use vars for user, group and locale suffix.
* Switch tests from `es_MX` to `es_ES`.
* Enable postgresql tests on freebsd.
* Work-arounds for service restart on freebsd.
* Fix git remote url change detection.
When comparing paths, unfrackgitpath must be called on both sides
of the comparison. Otherwise comparisons involving symlinks will
return incorrect results.
* Re-enable git test on OS X.
* Return None from unfrackgitpath when path is None.
Per a change in jinja2 2.9, local variables no longer are prefixed
with l_, so this updates AnsibleJ2Vars to pull in all locals (while
excluding some) regardless of name.
Fixes#20063
(cherry picked from commit 4d49b317929b86e1fc1b0cbace825ff73b372dc7)
* Fix `iterators` test for osx.
* Enable `git` test for osx.
* Enable `get_url` test for osx and freebsd.
* Remove unused SNI_URI from uri test.
* Use ansible hosted SNI host oustside docker.
* Enable `uri` test for osx and freebsd.
* Enable `gathering_facts` test for osx.
* Skip single task on osx until test is fixed.
* Add jinja2 groupby filter override to cast namedtuple to tuple. Fixes#20098
* Address some of the requested changes
* Quoting
* Print the python path and version
* Be less explicitly verbose, rely on implicit verbosity
* Adding support for Amazon ECR
This patch adds a new module named ecr, which can create, update or
destroy Amazon EC2 Container Registries. It also handles the management
of ECR policies.
* ecs_ecr: addressed review feeback
* Renaming ecr to ecs_ecr
* Fixed docs
* Removed bad doc about empty string handling
* Added example of `delete_policy`
* Removed `policy_text` option; switched policy to `json` type so
it can accept string or dict
* Added support for specifying registry_id
* Added explicit else after returned if clauses
* Added `force_set_policy` option
* Improved `set_repository_policy` error handling
* Fixed policy comparisons when AWS doesn't keep the ordering stable
* Moved `boto_exception` into the module
* Conditional include on ansible_network_os
* copy & paste error
* More tests
* More tests
* junos tests (based on vyos)
* remove excessive whitespace
* Pass in ansible_network_os
* net_command for ios
* consistent debug
* wrapp line
* ansible-test changes made in another PR
* ansible-test changes made in another PR
map + extract is the usual way to use it but map isn't available on
older versions of jinja2 that we still work with. Test extract even on
those versions.
* Only add Content-Type if not specified in headers. Fixes#20046
* Update documentation to indicate body_format will not override Content-Type if specified in headers
When using AWS we have to use the full domain name in the inventory file, which
we rather than the short name. This change avoids that ending up being
set in the tests.
* Cleanup git tests
* Split git tests in seperate files
* Remove use of repo_depth_url
* Use native yaml
* Remove unnecessary remote/local clones
* Fix newlines for yamllint
* If the hash is valid (full-length) but doesn't exist, git returns 128 instead of 1.
* Ensure git doesn't use hardlinks for shallow clones
* Reenable yum install root tests
No need for sos to test installroot. Something with less deps works
just as well.
* Fix yum installroot.
Fix module import to use fail_json when the modules aren't installed.
Remove wildcard imports
* Lsat task is supposed to remove sos so make that happen
* Add --installroot to YUM and DNF modules, issue #11310
This continues ansible-modules-core#1558, and
ansible-modules-core#1669
Allow specifying installroot for the yum and dnf modules
to install and remove packages in a location other than /.
* Remove empty aliases
* Simpler installroot set default logic
* Refactoring: split readkeys() into readfile() and parsekeys()
* Refactoring: split writekeys() into writefile() and serialize()
* authorized_key: support --diff
* Refactoring: remove no-longer used readkeys()/writekeys()
* Integration test for authorized_key in check mode
The gce_tag module can support updating tags on multiple instances via an instance_pattern field. Full Python regex is supported in the instance_pattern field.
'instance_pattern' and 'instance_name' are mutually exclusive and one must be specified.
The integration test for the gce_tag module has been updated to support the instance_pattern parameter. Unit tests have been added to test the list-manipulation functionality.
Run the integration test with:
TEST_FLAGS='--tags "test_gce_tag"' make gce
Run the unit tests with:
python test/units/modules/cloud/google/test_gce_tag.py
An inner single-quote pair breaks out of the outer single-quote
pair. Rather than escaping the inner quotes to protect against
this, just use the fact that `str()` is equivalent to `""`.
* Fix UnboundLocalError remote_head in git
Fixes#5505
The use of remote_head was a leftover of #4562.
remote_head is not necessary, since the repo is unchanged anyway and
after is set correctly.
Further changes:
* Set changed=True and msg once local_mods are detected and reset.
* Remove need_fetch that is always True (due to previous if) to improve
clarity
* Don't exit early for local_mods but run submodules update and
switch_version
* Add test for git with local modifications
* Use native yaml for apache2 test
* Test removal of default modules with force
a2enmod on debian has `-f`, but not on SUSE (runs there without force).
Therefore don't test that option on SUSE.
The docs already specify that the option is intended for Debian systems
only.
* Fix regression in jinja2 include search path
Since commit 3c39bb5, the 'ansible_search_path' variable is used to set
jinja2's search path for {% include %} directives. However, this path is
the the proper one because our templates live in 'templates' subdirs in
our search path.
This is a regression because previously, our include search path would
include the dirname of the currently interpreted file, which worked most
of the time.
fixes#18526
* Fix template lookup search path
Improve fix in commit c96c853 so that the search path contain both
template-suffixed paths as well as original paths.
ref PR #18617
* Add integration test for template lookups
Tests regression at #18526
This test fails on current devel branch and succeeds on PR #18617
* Add test cases for VyOS commands that don't honor paging settings
Testing for issue fixed in PR #18546
* Add provider line and fix indentation
For the way we invoke the tests we need to specify the `provider:`
Also fix the indentation on `register:`
This change corrects problems reported by the `yamllint` linter.
Since key duplication problems were removed in 4d48711, this commit
mainly fixes trailing spaces and extra empty lines at beginning/end of
files.
* Added test for sequenced-name instance generation (num_instances)
* Added param-check tags to tests that only do argument checking
Should be merged AFTER ansible/ansible-modules-core#4276
- Correct directory name in test/README.md
- Move code-smell tests to test/sanity/code-smell
- Update code-smell.sh to use new script paths
- Add test/integration/target-prefixes.win for ansible-test
- Move module unit tests to match module directory layout
* Network Test Documentation
Will need improving over time, though this ensure that everything that was in `ansible/test-network-modules` is in `ansible/ansible`
* Update README.md
* Inventory file
* Network module prefixes
In ansible-test we should skip tests for these modules, they will be
tested via another process.
* Update target-prefixes.network
If the facts returned by setup included strings that
had double quotes in them, the asserts in test_gathering_facts.yml
would fail with errors like:
"The conditional check '\"[{u'mounts': {u'options':
u'rw,context=\"system_u:\"'}}]\" != \"UNDEF_HW\"' failed. The error was:
template error while templating string: expected token 'end of statement
block', got 'system_u'. String: {% if \"[{u'mounts': {u'options':
u'rw,context=\"system_u:\"'}}]\" != \"UNDEF_HW\" %} True {% else %}
False {% endif %}"
For one example, if mount facts returned an 'options' field that
included double quoated selinux context ids, the test would fail.
Fix is removing the double quoting in the assert 'that:' lines,
and removing the unneeded double curly brackets.
* Add tag verification test (ansible-modules-core PR 2654)
* Fix typo
* Use smaller repo for testing, add dependency control
* Test is gpg exists before running git signing tasks
* Correct the test conditionals so that gpg1 is tested
Implement tag and skip_tag handling in the CLI() class. Change tag and
skip_tag command line options to be accepted multiple times on the CLI
and add them together rather than overwrite.
* Make it configurable whether to merge or overwrite multiple --tags arguments
* Make the base CLI class an abstractbaseclass so we can implement
functionality in parse() but still make subclasses implement it.
* Deprecate the overwrite feature of --tags with a message that the
default will change in 2.4 and go away in 2.5.
* Add documentation for merge_multiple_cli_flags
* Fix galaxy search so its tags argument does not conflict with generic tags
* Unit tests and more integration tests for tags
* Turn mount test back on
* Mount tests need PRIVILEGED so turn that back on
Revert "Revert "Set PRIVILEGED=true for non_destructive tests. (#17733)" (#17738)"
This reverts commit dc0fb1c212.
* Add a needs_privileged tag so that we can skip mount tests on centos6
Some containers timeout on shippable tests when run with privileged.
Unfortunately, some tests require that in order to run. Tagging those
allows us to skip those tests on the platforms that timeout when we get
ready to run the integration test in shippable.
* Centos6 times out with PRIVILEGED set so remove that (will disable the mount tests on centos6)
* Remove false start
* Add tests for the mount module
* Switch from unmounted to absent...
the code for mounting always modifies fstab so we need to always modify
fstab to cleanup as well.
* Fix comments and copyright
Includes:
* Tests to create, delete, resize, recreate and configure an Autoscaler
* Tests for parameters only. Use TEST_FLAGS="--tags 'param-check'" to execute only these tests.
Requires:
* An instance template to be configured in your project.
* Fixes to the controller text model
* Change command line args to text type
* Make display replace undecodable bytes with replacement chars. This
is only a problem on pyhton3 where surrogates can enter into the msg
but sys.stdout doesn't know how to handle them.
* Remove a deprecated playbook syntax in unicode.yml
* Fix up run_cmd to change its parameters to byte string at appropriate times.
The test_async test target was updated to accommodate changes in
output buffering behavior in python 3. This change in behavior
may need to be addressed in the future.
* New features for include_vars
include_vars.py now allows you to include an entire directory and its nested directories of variable files.
Added Features..
* Ignore by default *.md, *.py, and *.pyc
* Ignore any list of files.
* Only include files nested by depth (default=unlimited)
* Match only files matching (valid regex)
* Sort files alphabetically and load in that order.
* Sort directories alphabetically and load in that order.
```
- include_vars: 'vars/all.yml'
- name: include all.yml
include_vars:
file: 'vars/all.yml'
- name: include all yml files in vars/all and all nested directories
include_vars:
dir: 'vars/all'
- name: include all yml files in vars/all and all nested directories and save the output in test.
include_vars:
dir: 'vars/all'
name: test
- name: include all yml files in vars/services
include_vars:
dir: 'vars/services'
depth: 1
- name: include only bastion.yml files
include_vars:
dir: 'vars'
files_matching: 'bastion.yml'
- name: include only all yml files exception bastion.yml
include_vars:
dir: 'vars'
ignore_files: 'bastion.yml'
```
* Added whitelist for file extensisions (yaml, yml, json)
* Removed unit tests in favor of integration tests
While trying to fix the test suite on python3, I noticed
this test fail due to to_json adding more whitespace in
python3 than in python2. So -w should ignored those
differences.
Fixes for these are either rewriting to get rid of the need for the
functions or using six.moves to get equivalent functions for both
python2 and python3
Make !vault-encrypted create a AnsibleVaultUnicode
yaml object that can be used as a regular string object.
This allows a playbook to include a encrypted vault
blob for the value of a yaml variable. A 'secret_password'
variable can have it's value encrypted instead of having
to vault encrypt an entire vars file.
Add __ENCRYPTED__ to the vault yaml types so
template.Template can treat it similar
to __UNSAFE__ flags.
vault.VaultLib api changes:
- Split VaultLib.encrypt to encrypt and encrypt_bytestring
- VaultLib.encrypt() previously accepted the plaintext data
as either a byte string or a unicode string.
Doing the right thing based on the input type would fail
on py3 if given a arg of type 'bytes'. To simplify the
API, vaultlib.encrypt() now assumes input plaintext is a
py2 unicode or py3 str. It will encode to utf-8 then call
the new encrypt_bytestring(). The new methods are less
ambiguous.
- moved VaultLib.is_encrypted logic to vault module scope
and split to is_encrypted() and is_encrypted_file().
Add a test/unit/mock/yaml_helper.py
It has some helpers for testing parsing/yaml
Integration tests added as roles test_vault and test_vault_embedded
* actions/unarchive: fix unarchive from remote url
Currently unarchive from remote url does not work because the core
unarchive module was updated to support 'remote_src' [1], but the
unarchive action plugin was not updated for this. This causes failures
because the action plugin assumes it needs to copy a file to the
remote server, but in the case of downloading a file from a remote
url a local file does not exist, so an error occurs when the file is
not found.
[1] https://github.com/ansible/ansible-modules-core/commit/467516e
* test_unarchive: fix test with wrong remote_src use
The non-ascii filenames test had improperly set remote_src=yes even
though it was actually copying the file from the local machine (i.e.
the file did not already exist remotely). This test was passing
until the remote_src behavior of unarchive was fixed in 276550f.
Rather than repeatedly searching for tasks by uuid via iterating over
all known blocks, cache the tasks when they are added to the PlayIterator
so the lookup becomes a simple key check in a dict.
* Query lookup plugin
* Add license and docstrings
* Add python3-ish imports
* Change query plugin type from lookup to filter
* Switch from dq to jsonpath_rw
* Add integration test for query filter
* Rename query filter to json_query
* Add jsonpath-rw
* Rename query filter to json_query
* Switch query implementation from jsonpath-rw to jmespath
When a task result has an empty results list, the
list should be ignored when determining the results
of `_check_key`. Here the empty list is treated the
same as a non-existent list.
This fixes a bug that manifests itself with squashed
items - namely the task result contains the correct
value for the key, but an empty results list. The
empty results list was treated as zero failures
when deciding which handler to call - so the task
show as a success in the output, but is deemed to
have failed when deciding whether to continue.
This also demonstrates a mismatch between task
result processing and play iteration.
A test is also added for this case, but it would not
have caught the bug - because the bug is really in
the display, and not the success/failure of the
task (visually the test is more accurate).
Fixesansible/ansible-modules-core#4214
Binary module tests now download pre-compiled binaries from S3
instead of downloading go and compiling the modules.
Files downloaded form SourceForge are now downloaded from S3.
* Enabled more tests in test_service on systems with systemd.
* Fix inconsistency in cleaning up test service. The conditions for service setup and service cleanup were different.
* Add check mode run for service enable.
Enable color output from integration tests in Docker containers:
* In run_tests.sh when output is attached to a terminal.
* In shippable/integration.sh using force mode (can be disabled).
Also fix blocks tests to work with or without color output
* add check_mode option for tasks
includes example testcases for the template module
* extend check_mode option
* replace always_run, see also proposal rename_always_run
* rename always_run where used and add deprecation warning
* add some documentation
* have check_mode overwrite always_run
* use unique template name to prevent conflicts
test_check_mode was right before, but failed due to using the same filename as other roles
* still mention always_run in the docs
* set deprecation of always_run to version 2.4
* fix rst style
* expand documentation on per-task check mode
I suspect this problem was masked previously as older versions of pywinrm
where significantly slower, allowing more time for the windows installer service to complete.
6eefc11c converted task.loop_control into an object, but while the other
callers were updated to use .loop_var instead of .get('loop_var'), this
site was overlooked.
This can be reproduced by including with loop_control a file that does
set_fact; a simple regression test along these lines is included.
* Lookup unencrypted password must not include salt
* Integration test lookup: remove previous directory
* Test that lookup password doesn't return salt
* Lookup password: test behavior with empty encrypt parameter
Closes#16189
* add git shallow fetch test
covers https://github.com/ansible/ansible-modules-core/issues/3782
updating a repo with depth=1 fails silently if version==HEAD
* disable git shallow tests for old git versions
Older git versions don't treat the --depth option correctly.
While the git module tried to work around this and introduced subtle
bugs, ansible/ansible-modules-core#3794 falls back to full checkouts.
Don't run the tests then.
* reduce async sleep time in test
* make zypper test less destructive (don't break following uses of zypper)
* fix ca cert on suse
* fix/enable postgres/mysql on opensuseleap
* fix mysql test for mysql versions 5.7.6 and newer
* skip sni_host check on ubuntu1604
* add HTTPTESTER flag for test_uri
ubuntu 16.04 uses dash which drops env variables containing a dot
we work around this by adding an explicit env variable to enable httptester
* alias module is very basic and removing it leads to the suse default
config failing
* future improvements might test different modules and the effect of
them being removed
* When var name is the same as var content, try to template it before reporting that var is not defined.
Add asserts in test_var_blending to check this special corner case.
* Fix integration tests when using debug with list or dict.
* extend zypper integration tests
* fix caching issue of local RPMS on openSUSE 42.1 Leap
* add tests for simultaneous install/remove via prefixes +-
* test fail cases (rm patch or URL)
* test patch install (succes, unchanged second run, fail on wrong name)
* add test for pattern install
* include options refresh and priority introduced in ansible/ansible-modules-extras#1990
* test repo update both based on name and URL
* run on all suse flavors
* test overwriting repos by url or name
* add integration test for apache2_module
* fix assert syntax
* check os_family for this test
* add comment why only debian and suse are tested
* move apache2 test to destructive
* install apache2 package
- get_real_file will decrypt vault encrypted files and return a path to
a temporary file.
- cleanup_real_file will remove a temporary file created previously with
get_real_file
* Ziploader proof of concept (jimi-c)
* Cleanups to proof of concept ziploader branch:
* python3 compatible base64 encoding
* zipfile compression (still need to enable toggling this off for
systems without zlib support in python)
* Allow non-wildcard imports (still need to make this recusrsive so that
we can have module_utils code that imports other module_utils code.)
* Better tracebacks: module filename is kept and module_utils directory
is kept so that tracebacks show the real filenames that the errors
appear in.
* Make sure we import modules that are used into the module_utils files that they are used in.
* Set ansible version in a more pythonic way for ziploader than we were doing in module replacer
* Make it possible to set the module compression as an inventory var
This may be necessary on systems where python has been compiled without
zlib compression.
* Refactoring of module_common code:
* module replacer only replaces values that make sense for that type of
file (example: don't attempt to replace python imports if we're in
a powershell module).
* Implement configurable shebang support for ziploader wrapper
* Implement client-side constants (for SELINUX_SPECIAL_FS and SYSLOG)
via environment variable.
* Remove strip_comments param as we're never going to use it (ruins line
numbering)
* Don't repeat ourselves about detecting REPLACER
* Add an easy way to debug
* Port test-module to the ziploader-aware modify_module()
* strip comments and blank lines from the wrapper so we send less over the wire.
* Comments cleanup
* Remember to output write the module line itself in powershell modules
* for line in lines strips the newlines so we have to add them back in
* Unit tests exposed a problem where nested blocks did not correctly
hit rescue/always portions of parent blocks
* Cleaned up logic in PlayIterator
* Unfortunately fixing the above exposed a potential problem in the
block integration tests, where a failure in an "always" section may
always lead to a failed state and the termination of execution
beyond that point, so certain parts of the block integration test
were disabled.
* fetch_url shouldn't both accept follow_redirects and support follow_redircts via module.params
* Default follow_redirects for open_url should be 'urllib2'
* Add redirect test for get_url
The dep chain for roles created during the compile step had bugs, in
which the dep chain was overwriten and the original tasks in the role
were not assigned a dep chain. This lead to problems in determining
whether roles had already run when in a "diamond" structure, and in
some cases roles were not correctly getting variables from parents.
Fixes#14046
Also adds a new var, used by the prepare_tests role, to prevent it from
deleting the temp test directory at the start of each play to avoid any
potential race conditions
At its most basic, this is nothing more than an array or hash lookup,
but when used in conjunction with map, it is very useful. For example,
while constructing an "ssh-keyscan …" command to update known_hosts on
all hosts in a group, one can get a list of IP addresses with:
groups['x']|map('extract', hostvars, 'ec2_ip_address')|list
This returns hostvars[a].ec2_ip_address, hostvars[b].ec2_ip_address, and
so on. You can even specify an array of keys for a recursive lookup, and
mix string and integer keys depending on what you're looking up:
['localhost']|map('extract', hostvars, ['vars','group_names',0])|first
== hostvars['localhost']['vars']['group_names'][0]
== 'ungrouped'
Includes documentation and tests.
Modeled after the yum tests but also tests local package installations
as fixed with PR#1256.
This depends on PRs #1256, #1261 and #1262 in ansible-modules-extra.
Looks like there are two pattern caches that need to be cleared for this to work- added the second one.
Added integration tests for add_host to prevent future regressions.
Check for database presence in a nice way, rather than
dropping the database.
Thankfully there was a syntax error in the previous
version, so no database would have been dropped.
There was no check for whether it succeeded or not.
* Fix the task_vars parameter to not default to a mutable type (dict)
* Implement invocation in the base class's run() method have each action
module call the run() method's implemention in the base class.
* Return values from the action plugins' run() method takes the return
value from the base class run() method into account so that invocation
makes its way to the output.
Fixes#12869
Revert "Remove auto-added invocation return value as it is not used by v2 and could leak sensitive data."
This reverts commit 6ce6b20268.
Remove the note that invocation was removed as we've now restored it.
Revert "keyword not in ubuntu 14.04"
This reverts commit 5c01622457.
Revert "remove invocation keyword check"
This reverts commit 5177cb3f74.
ansible-playbook now works when run with a playbook
that includes a role that includes another role
specified using csv format
Updated one of the roles used in the tests to fix
broken tests - `make test_galaxy` now works
Fixes#11486. Also addresses the problem alluded to in #10620.
better error reporting on fetching errors
use scm if it exists over src
unified functions in requirements
simplified logic
added verbose to tests
cleanup code refs, unused options and dead code
moved get_opt to base class
fixes#11920fixes#12612fixes#10454
corrected output from default callback
added new tests for no_log loops
updated makefile test to check for both positive and negative occurrences of no_log
* Add exception handling when running PowerShell modules to provide exception message and stack trace.
* Enable strict mode for all PowerShell modules and internal commands.
* Update common PowerShell code to fix strict mode errors.
* Fix an issue with Set-Attr where it would not replace an existing property if already set.
* Add tests for exception handling using modified win_ping modules.
1. The test did "name: '{{hostnames}}.{{item}}'" inside a with_sequence
loop, which didn't do what was intended: it expanded hostnames into
an array, appended ".1", and set name to the resulting string. This
can be converted to a simple with_items loop.
2. Some of the entries in hostnames contained punctuation characters,
which I see no reason to support in inventory hostnames anyway.
3. Once the add_host failures are fixed, the playbook later fails when
the unicode hostnames are interpolated into debug output in ssh.py
due to an encoding error. This is only one of the many places that
may fail when using unicode inventory hostnames; we work around it
by providing an ansible_ssh_host setting.
`assert (condition, message)` gets parsed by Python as `assert
a_two_tuple`, and a 2-element tuple is never False.
Discovered by compileall on Python 3.4, which emits a SyntaxWarning for
this common mistake.
* Add exception handling when running PowerShell modules to provide exception message and stack trace.
* Enable strict mode for all PowerShell modules and internal commands.
* Update common PowerShell code to fix strict mode errors.
* Fix an issue with Set-Attr where it would not replace an existing property if already set.
* Add tests for exception handling using modified win_ping modules.
This change is similar to https://github.com/ansible/ansible/pull/10465
It extends the logic there to also support none types. Right now if you have
a '!!null' in yaml, and that var gets passed around, it will get converted to
a string.
eg. defaults/main.yml
```
ENABLE_AWESOME_FEATURE: !!null # Yaml Null
OTHER_CONFIG:
secret1: "so_secret"
secret2: "even_more_secret"
CONFIG:
hostname: "some_hostname"
features:
awesame_feature: "{{ ENABLE_AWESOME_FEATURE}}"
secrets: "{{ OTHER_CONFIG }}"
```
If you output `CONFIG` to json or yaml, the feature flag would get represented in the output
as a string instead of as a null, but secrets would get represented as a dictionary. This is
a mis-match in behaviour where some "types" are retained and others are not. This change
should fix the issue.
I also updated the template test to test for this and made the changes to v2.
Added a changelog entry specifically for the change from empty string to null as the default.
Made the null representation configurable.
It still defaults to the python NoneType but can be overriden to be an emptystring by updating
the DEFAULT_NULL_REPRESENTATION config.
- Add support for inserting module args into PowerShell modules. Fixes#11661.
- Support Windows paths containing spaces. Applies changes from #10727 to v2. Fixes#9999. Should also fixansible/ansible-modules-core#944 and ansible/ansible-modules-core#1007.
- Change how execution policy is set for running remote scripts. Applies changes from #11092 to v2. Also fixesansible/ansible-modules-core#1776.
- Use codepage 65001 (UTF-8) for WinRM connection instead of default (CP437), convert command to UTF-8 and results from UTF-8. Replaces changes from #10024. Fixes#11198.
- Close WinRM connection when task completes.
- Use win_stat, win_file and win_copy modules instead of stat, file and copy when called from within other action plugins (only when using WinRM+PowerShell).
- Unquote Windows path arguments before passing to win_stat, win_file, win_copy and slurp modules (only when using WinRM/PowerShell).
- Check for win_ping module to determine if core modules are missing (only when using WinRM/PowerShell).
- Add stdout_lines to result from running low level commands (so stdout_lines is available when using raw/script).
- Update copy action plugin to use shell functions for joining paths and checking for trailing slash.
- Update fetch action plugin to unquote source path when using Windows paths.
- Add win_copy and win_template action plugins that inherit from copy and template.
- Support running .bat and .cmd scripts using default system encoding instead of UTF-8.
- Always send PowerShell commands as base64-encoded blobs to allow for running simple PowerShell commands via raw.
- Support running modules on Windows with interpreters other than PowerShell.
- Update integration tests to support above changes and test unicode fixes.
- Add test for win_user error from ansible/ansible-modules-core#1241 (fixed by ansible/ansible-modules-core#1774).
- Add test for additional win_stat output values (implemented by ansible/ansible-modules-core#1473).
- Add test for OS architecture and name from setup.ps1 (implemented by ansible/ansible-modules-core#1100).
All WinRM integration tests pass for me with these changes.
* Perhaps the only precedence change, in 2.0+ variables from set_fact
will not override params to an include file, as params are expected
to be more specific than host-based variables.
* Uncommented long-form include example.
In issue #10073 a misbehaviour in literal handling for inline lookup
arguments that can cause unexpected behaviur was reported. This
integration testcase reproduce the problem.
After applying pull request #10991 the issue is fixed and the test
passes.
Signed-off-by: Jan Losinski <losinski@wh2.tu-dresden.de>
Before this change if a variable was of type int or bool and the variable was referenced
by another variable, the type would change to string.
eg. defaults/main.yml
```
PORT: 4567
OTHER_CONFIG:
secret1: "so_secret"
secret2: "even_more_secret"
CONFIG:
hostname: "some_hostname"
port: "{{ PORT }}"
secrets: "{{ OTHER_CONFIG }}"
```
If you output `CONFIG` to json or yaml, the port would get represented in the output as a
string instead of as a number, but secrets would get represented as a dictionary. This is
a mis-match in behaviour where some "types" are retained and others are not. This change
should fix the issue.
Update template test to also test var retainment.
Make the template changes in v2.
Update to only short-circuit for booleans and numbers.
Added an entry to the changelog.
The --force-handlers command line argument was not correctly running
handlers on hosts which had tasks that later failed. This corrects that,
and also allows you to specify force_handlers in ansible.cfg or in a
play.
- become constants inherit existing sudo/su ones
- become command line options, marked sudo/su as deprecated and moved sudo/su passwords to runas group
- changed method signatures as privlege escalation is collapsed to become
- added tests for su and become, diabled su for lack of support in local.py
- updated playbook,play and task objects to become
- added become to runner
- added whoami test for become/sudo/su
- added home override dir for plugins
- removed useless method from ask pass
- forced become pass to always be string also uses to_bytes
- fixed fakerunner for tests
- corrected reference in synchronize action plugin
- added pfexec (needs testing)
- removed unused sudo/su in runner init
- removed deprecated info
- updated pe tests to allow to run under sudo and not need root
- normalized become options into a funciton to avoid duplication and inconsistencies
- pushed suppored list to connection classs property
- updated all connection plugins to latest 'become' pe
- includes fixes from feedback (including typos)
- added draft docs
- stub of become_exe, leaving for future v2 fixes
Instead of getting rid of pipes.quote, use pipes.quote and get rid of
the manually entered toplevel quotes. This should properly escape
backslashes, quotes, and other characters.
Also fix the new checksumming python "one-liner" for csh.
ansible_shell_type needs to be set to csh.
Fixes#10363Fixes#10353
When using the "creates" option with the unarchive module, set changed
to False if the file already exists. This behavior is consistent with
other modules which use "creates", such as command and shell.
Submission includes support for
- creating and registering services and checks
- reading, writing and lookup for values in consul's kv store
- creating and manipulating sessions for distributed locking on values in the kv
- creating and manipulating ACLs for restricting access to the kv store
- inventory support that reads the Consul catalog and group nodes according to
- datacenters
- exposed services
- service availability
- arbitrary groupings from the kv store
This submission makes extensive use of the python-consul library and this is required
as a dependency and can be installed from pip.
The tests were written to target a vagrant cluster which can be setup by following the
instructions here http://github.com/sgargan/consul-vagrant
The test template output is compared against an expected output. Since
hostvars contains information specific to the host it runs on we can't
test the python2.6 bug this way.
This reverts commit af4d6105a5.
Now uses sha1 checksums following merge of 9688.
Also I undid the changes I made to fetch.py
win_template.py now uses conn.shell.has_trailing_slash and
conn.shell.join_path
updated integration tests.
(and a few unicode tests because they make clearer what we're testing)
works. disabled for now since they mostly do not pass on 1.8. We'll
want to enable them when we start integration testing the v2 tree.
Fixes a case where the variable 'foo' may exist, but the with_items loop
was used on something like 'foo.results', where 'results' was not a valid
attribute of 'foo'. Prior to this patch, conditionals were not evaluated
until later, meaning there was no opportunity to allow a test to skip the
task or item based on it being undefined.
This commit also makes the dest parameter optional, unless update=yes
(the default), since it is not required for queries without an update.
Fixes#8630
Also modifies the template action plugin to use this new param
when executing the file/copy modules for templating so that links
are preserved correctly.
Fixes#8998
Generate warnings when users are shelling out to commands
rather than using modules
Can be turned off on a per-action line with the documented
warn=False flag. Can be turned off globally using
command_warnings = False in ansible config file.
Print out warnings using the standard playbook callbacks.
Created some additional tests in TestRunner.test_command
and also a demonstration playbook.
Older git archive commands create tar archives even with a tar.gz
extension. So change it to always create tar archives and have
the install_role method cope.
Removed ssh roles from the test case as they don't work unless
you can connect to bitbucket via ssh and have your key there.
Corrected a minor typo in error messages
* Roles can now be given a friendly name as third field in role spec csv
* Roles can be installed from URL (not just from archived SCMs)
* Integration tests to demonstrate this
* Unit tests to ensure that role spec parsing works as expected
Use mysql_variable to query, set and update variables.
Assert using user and password to query, set and update variables.
Assert using single quotes, double quotes and no quotes when using variables
Use mysql_user module to create, delete users.
Update user password and ensure new password was updated for the correct user.
Assert user has access to multiple databases
Assert user creation, deleting using different user privilege and ensure privilege work correctly.
* Strip lookup calls out of inventory variables and clean unsafe data
returned from lookup plugins (CVE-2014-4966)
* Make sure vars don't insert extra parameters into module args and prevent
duplicate params from superseding previous params (CVE-2014-4967)
Adding the following Test Coverage for RedHat and Debian OS:
- Use mysql_db module to create, delete databases using different encoding.
- Backup/Restore databases using different file format: sql, gz and bz2.
- Create and delete databases using different user privilege.
- Assert database creation, deleting and content using system commands.
Looks like #6881 is also affected by this bug.
Currently when using lineinfile insertafter=EOF for a file that has no newline then the added line is not prefixed with a newline.
As a concrete example, ubuntu distro default /etc/ssh/sshd_config has no newline so lineinfile module cannot be used to add a new line to the end of the file.
Remove 'python-apt' before testing the 'apt_repository' module. The
'apt_repository' module has the smarts to auto-install python-apt. The
integration test will now assert the correct auto-install behavior.
* the current state of the ELB was not reflected properly when checking
the status after a change was made.
* invalid zones caused a traceback when enabling/disabling zones
Changes include:
* Update Makefile to use credentials.yml when it exists
* Add details on the use of the credentials.yml file to README.md.
* Update credentials.template comments
The `apt` and `yum` modules will automatically install python dependencies.
This change updates the existing integration tests to test whether auto-install
of dependencies is functioning properly.
- unified set attribute functions ... not sure why 2 identical functions
exist with diff names, now there are 3 while i repoint all modules to 1
- fixed issue with symlinks being created w/o existing src when force=no
- refactored conditionals, simplified where possible
- added tests for symlink to nonexistant source, with both force options
- made symlink on existing attomic (force)
Tests several ways to specify the repository. For every repo added, the test
asserts that:
* the apt-cache was updated as expected (depends on `update_cache` parameter)
* the PPA key was installed (depends on `repo` format)
To support parallel cloud test execution, create and provide a random string to
cloud integration tests. The variable 'resource_prefix' can be used in cloud
roles and during resource cleanup to safely create/destroy cloud-based
resources.
Additional changes include:
* The roles test_ec2_key and test_ec2_group were updated to use to
{{resource_prefix}}.
* Additionally, the Makefile was updated to set resource_prefix to a random
string. The Makefile will also use 'resource_prefix' during cloud_cleanup.
* All test_ec2* roles were updated to add 'setup_ec2' as a role dependency.
The 'service' utility was unable to find the 'ansible_test' service due to an
unexpected filename. This patch corrects the filename and adjusts the
permissions to match other service scripts within /etc/init/.
