Merge "Remove the transaction scope from task executions API"

mistral/api/controllers/v2/task.py

@@ -102,9 +102,8 @@ def _get_task_resources_with_results(wf_ex_id=None):
     if wf_ex_id:
         filters['workflow_execution_id'] = wf_ex_id
 
-    with db_api.transaction():
-        task_exs = db_api.get_task_executions(**filters)
-        tasks = [_get_task_resource_with_result(t_e) for t_e in task_exs]
+    task_exs = db_api.get_task_executions(**filters)
+    tasks = [_get_task_resource_with_result(t_e) for t_e in task_exs]
 
     return Tasks(tasks=tasks)
 

mistral/engine/policies.py

@@ -272,6 +272,14 @@ class RetryPolicy(base.TaskPolicy):
         """
         super(RetryPolicy, self).after_task_complete(task_ex, task_spec)
 
+        # TODO(m4dcoder): If the task_ex.executions collection is not called,
+        # then the retry_no in the runtime_context of the task_ex will not
+        # be updated accurately. To be exact, the retry_no will be one
+        # iteration behind. task_ex.executions was originally called in
+        # get_task_execution_result but it was refactored to use
+        # db_api.get_action_executions to support session-less use cases.
+        action_ex = task_ex.executions  # noqa
+
         context_key = 'retry_task_policy'
 
         runtime_context = _ensure_context_has_key(

mistral/tests/unit/engine/test_dataflow.py

@@ -13,6 +13,8 @@
 #    See the License for the specific language governing permissions and
 #    limitations under the License.
 
+import mock
+
 from oslo_config import cfg
 from oslo_log import log as logging
 
@@ -472,26 +474,36 @@ class DataFlowTest(test_base.BaseTest):
             }
         )
 
-        task_ex.executions.append(models.ActionExecution(
+        action_exs = []
+
+        action_exs.append(models.ActionExecution(
             name='my_action',
             output={'result': 1},
             accepted=True,
             runtime_context={'with_items_index': 0}
         ))
 
-        self.assertEqual([1], data_flow.get_task_execution_result(task_ex))
+        with mock.patch.object(db_api, 'get_action_executions',
+                               return_value=action_exs):
+            self.assertEqual([1], data_flow.get_task_execution_result(task_ex))
 
-        task_ex.executions.append(models.ActionExecution(
+        action_exs.append(models.ActionExecution(
             name='my_action',
             output={'result': 1},
             accepted=True,
             runtime_context={'with_items_index': 0}
         ))
-        task_ex.executions.append(models.ActionExecution(
+
+        action_exs.append(models.ActionExecution(
             name='my_action',
             output={'result': 1},
             accepted=False,
             runtime_context={'with_items_index': 0}
         ))
 
-        self.assertEqual([1, 1], data_flow.get_task_execution_result(task_ex))
+        with mock.patch.object(db_api, 'get_action_executions',
+                               return_value=action_exs):
+            self.assertEqual(
+                [1, 1],
+                data_flow.get_task_execution_result(task_ex)
+            )

mistral/tests/unit/engine/test_policies.py

@@ -715,8 +715,8 @@ class PoliciesTest(base.EngineTestCase):
         wf_ex = db_api.get_workflow_execution(wf_ex.id)
         task_ex = wf_ex.task_executions[0]
 
-        self.assertEqual(
-            {},
+        self.assertDictEqual(
+            {'retry_no': 1},
             task_ex.runtime_context['retry_task_policy']
         )
 

mistral/workflow/data_flow.py

@@ -87,14 +87,20 @@ def invalidate_task_execution_result(task_ex):
 
 
 def get_task_execution_result(task_ex):
-    action_execs = task_ex.executions
+    # Use of task_ex.executions requires a session to lazy load the action
+    # executions. This get_task_execution_result method is also invoked
+    # from get_all in the task execution API controller. If there is a lot of
+    # read against the API, it will lead to a lot of unnecessary DB locks
+    # which result in possible deadlocks and WF execution failures. Therefore,
+    # use db_api.get_action_executions here to avoid session-less use cases.
+    action_execs = db_api.get_action_executions(task_execution_id=task_ex.id)
     action_execs.sort(
         key=lambda x: x.runtime_context.get('with_items_index')
     )
 
     results = [
         _extract_execution_result(ex)
-        for ex in task_ex.executions
+        for ex in action_execs
         if hasattr(ex, 'output') and ex.accepted
     ]