Commit graph

46 commits

Author SHA1 Message Date
James Cammarata
f4ecbc4c3f Fixing iterator bug related to reworking of end-of-role detection
Bug was introduced in cae682607

(cherry picked from commit 9d549c3498)
2017-01-19 10:11:52 -06:00
James Cammarata
b12256acf2 Reworking the way end of role detection is done
Rather than trying to enumerate tasks or track an ever changing cur_role
flag in PlayIterator, this change simply sets a flag on the last block in
the list of blocks returned by Role.compile(). The PlayIterator then checks
for that flag when the cur_block number is incremented, and marks the role
as complete if the given host had any tasks run in that role.

Fixes #20224

(cherry picked from commit cae682607c)
2017-01-19 10:11:44 -06:00
Steve Kuznetsov
e13f3e3c07 Change v2_playbook_on_start logic to positively detect legacy plugins
In order to support legacy plugins, the following two method signatures
are allowed for `CallbackBase.v2_playbook_on_start`:

def v2_playbook_on_start(self):
def v2_playbook_on_start(self, playbook):

Previously, the logic to handle this divergence checked to see if the
callback plugin being called supported an argument named `playbook`
in its `v2_playbook_on_start` method. This was fragile in a few ways:
 - if a plugin author did not use the literal `playbook` to name their
   method argument, their plugin would not be called correctly
 - if a plugin author wrapped their `v2_playbook_on_start` method and
   by doing so changed the argspec to no longer expose an argument
   with that literal name, their plugin would not be called correctly

In order to continue to support both types of callback for backwards
compatibility while making the call more robust for plugin authors,
the logic can be reversed in order to have a positive check for the old
method signature instead of a positive check for the new one.

Signed-off-by: Steve Kuznetsov <skuznets@redhat.com>
(cherry picked from commit 0bc35354ce)
2016-11-01 07:53:04 -07:00
James Cammarata
47acf55fa9 Cache tasks by uuid in PlayIterator for O(1) lookups
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.
2016-08-10 12:49:19 -05:00
James Cammarata
e244895174 Avoid copying task parents in TaskExecutor
As Block.copy() is potentially expensive, avoid copying the parent
structure of tasks in TaskExecutor.
2016-08-08 15:58:46 -05:00
James Cammarata
06d4f4ad0e Move tasks/blocks to a single parent model 2016-08-08 15:58:46 -05:00
James Cammarata
d2b3b2c03e Performance improvements 2016-08-08 15:58:46 -05:00
James Cammarata
159aa26b36 FEATURE: adding variable serial batches
This feature changes the scalar value of `serial:` to a list, which
allows users to specify a list of values, so batches can be ramped
up (commonly called "canary" setups):

- hosts: all
  serial: [1, 5, 10, "100%"]
  tasks:
  ...
2016-08-04 15:04:10 -05:00
Matt Clay
292785ff2b Parse async response in async action. (#16534)
* Parse async response in async action.
* Add async test for non-JSON data before module output.
* Fix existing async unit test.

Resolves #16156
2016-07-01 14:52:45 -07:00
Toshio Kuratomi
f86df7c88b Test that task.args are also set properly (#15950) 2016-06-01 13:28:57 -07:00
jctanner
373b23cc24 Fix mock loader for osx /etc symlinks (#16074)
Fix role based unit tests for osx via mock.patch
2016-05-31 18:02:39 -04:00
Matt Davis
bc7405efc9 Merge pull request #15797 from robinro/patch-1
reduce async timeout in unittests
2016-05-12 17:19:10 -07:00
Toshio Kuratomi
292f0ed0d6 If we can't squash for any reason, then simply do not optimize the items loop.
Also add more squashing testcases

Fixes #15649
2016-05-12 10:06:18 -07:00
Robin Roth
254cf9ea68 reduce async timeout
reduce from 3 sec to 0.1 sec; the 3 sec sleep was about half of the total unittest time on my development machine...
2016-05-10 08:09:28 +02:00
Toshio Kuratomi
5c7ad654db Add some more tests for item squashing 2016-05-05 20:33:44 -07:00
nitzmahone
133395db30 add jimi-c's unit test for squashed skip results, tweaked is_skipped() logic to pass 2016-05-05 15:29:10 -07:00
James Cammarata
6eefc11c39 Make the loop variable (item by default) settable per task
Required for include+with* tasks which may include files that also
have tasks containing a with* loop.

Fixes #12736
2016-04-19 01:42:40 -04:00
Jonathon Klobucar
7bee994e1c Fix for serial when percent amount is less than one host (#15396)
Ansible when there was a percentage that was calculated to be less than
1.0 would run all hosts as the value for a rolling update.

The error is due to the fact that Python will round a
float that is under 1.0 to 0, which will trigger the case of
0 hosts. The 0 host case tells ansible to run all hosts.

The fix will see if the percentage calculation after int
conversion is 0 and will else to 1 host.
2016-04-13 10:49:38 -04:00
James Cammarata
81788e627d Adding unit tests for TaskResult 2016-03-14 01:33:46 -04:00
James Cammarata
9d61a6cba8 Fixing PlayIterator bugs
* 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.
2016-03-09 13:31:30 -05:00
James Cammarata
299d93f6e9 Updating unit tests for PlayIterator
This knowingly introduces a broken test, planning to fix that later.
2016-03-07 13:03:06 -05:00
James Cammarata
e02b98274b issue callbacks per item and retry fails
- now workers passes queue to task_executor so it can send back events per item and on retry attempt
- updated result class to pass along events to strategy
- base strategy updated to forward new events to callback
- callbacks now remove 'items' on final result but process them directly when invoked per item
- new callback method to deal with retry attempt messages (also now obeys nolog)
- updated tests to match new signature of task_executor

fixes #14558
fixes #14072
2016-02-23 15:07:06 -05:00
Toshio Kuratomi
0e410bbc8a Squashing was occuring even though pkgs didn't have a template that would be affected by squash
This broke other uses of looping (looping for delegate_to in the
reported bug)

Fixes #13980
2016-01-27 21:55:00 -08:00
Toshio Kuratomi
318bfbb207 Migrate cli and dependencies to use global display 2015-11-11 10:44:22 -08:00
Toshio Kuratomi
aa034f4fcf Add unit tests for issue #12976 2015-11-05 10:52:37 -08:00
Toshio Kuratomi
37052a983a Fix unittest dailure with the new squash items code 2015-11-05 10:45:54 -08:00
James Cammarata
0e1a5919c8 Adding unit tests for PlaybookExecutor _get_serialized_batches 2015-10-12 11:27:37 -04:00
James Cammarata
dfa33d0f23 Tweak variable manager use in role includes to avoid test failures 2015-09-30 08:05:55 -04:00
Marius Gedminas
c22f4ee697 Add a comment explaining why async = 1 in the test 2015-09-07 19:46:05 +03:00
Marius Gedminas
1a1c9203e2 Python 3: fix TypeError: unorderable types in test
The full error was

======================================================================
ERROR: test_task_executor_execute (units.executor.test_task_executor.TestTaskExecutor)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mg/src/ansible/test/units/executor/test_task_executor.py", line 252, in test_task_executor_execute
    mock_action.run.return_value = dict(ansible_facts=dict())
  File "/home/mg/src/ansible/lib/ansible/executor/task_executor.py", line 317, in _execute
    if self._task.async > 0:
TypeError: unorderable types: MagicMock() > int()

----------------------------------------------------------------------

Experiments show that Python 2 MagicMock() > 0 is true, so I'm setting
the async property on mock_task to 1.  (If I set it to 0, the test fails
anyway.)
2015-09-07 09:49:28 +03:00
James Cammarata
b2cb3153f1 Fix unit tests for PlayIterator changes regarding fact gathering 2015-09-03 14:25:43 -04:00
James Cammarata
ccaf47bd97 Fix unit tests broken by 057712c1 2015-08-28 16:44:04 -04:00
James Cammarata
9891bcbcae Fix unit tests broken by previous commit 2015-08-04 12:40:13 -04:00
James Cammarata
993ce592b1 Update unit tests for flush_handlers fix 2015-07-22 11:30:05 -04:00
James Cammarata
e64989beb4 Moving ConnectionInformation -> PlayContext
Also making PlayContext a child class of the Playbook Base class,
which gives it access to all of the FieldAttribute code to ensure
field values are correctly typed after post_validation

Fixes #11381
2015-07-21 12:13:50 -04:00
James Cammarata
3d77723e3d Cleaning up some stuff related to the connection info/become/executable change 2015-07-19 01:12:23 -04:00
Brian Coca
323362e23a added stdout to test result 2015-07-14 15:59:18 -04:00
Brian Coca
aa6486778f fixed become test to match new expected output 2015-07-06 15:34:38 -04:00
Brian Coca
5bac17de51 fixed pfexec test 2015-06-15 01:20:38 -04:00
Matt Martz
230be812ba Don't test for play.become_pass any longer 2015-06-06 09:25:30 -05:00
James Cammarata
d4a31e8d26 Adding unit tests for TaskExecutor (v2) 2015-05-24 07:47:06 -05:00
James Cammarata
838ff32001 Fix unit test for conn_info (v2)
The default user expected in the connection information is the current
user, not root
2015-05-22 08:38:39 -05:00
James Cammarata
ecd5eb902d Adding unit tests for ConnectionInformation (v2) 2015-05-22 03:32:40 -05:00
James Cammarata
1ca8cb8553 Fixing up v2 unit tests 2015-05-12 12:24:57 -05:00
Matt Martz
a0fc8bb0bd Testing additions and fixes
* Fix import pathing for units.mock
* Add some additional requirements
* Use compileall to test compatiblity with different python versions
2015-05-08 13:40:02 -05:00
James Cammarata
ce3ef7f4c1 Making the switch to v2 2015-05-03 21:47:26 -05:00