Wait for the policy to be done in tests
As observed in the amqp job, we sometimes signal scale policy too fast,
without waiting for the medata to be set. It creates a timeout. We can
check the metadata of the policy before signaling again, to make sure
that we can move forward.
Change-Id: I9857803ef960efbd034f05985ec8fcc7272e2f70
diff --git a/common/test.py b/common/test.py
index f52cda3..f2e7dd7 100644
--- a/common/test.py
+++ b/common/test.py
@@ -655,10 +655,14 @@
return matched
time.sleep(build_interval)
- def check_autoscale_complete(self, stack_id, expected_num):
+ def check_autoscale_complete(self, stack_id, expected_num, parent_stack,
+ policy):
res_list = self.client.resources.list(stack_id)
all_res_complete = all(res.resource_status in ('UPDATE_COMPLETE',
'CREATE_COMPLETE')
for res in res_list)
all_res = len(res_list) == expected_num
- return all_res and all_res_complete
+ if all_res and all_res_complete:
+ metadata = self.client.resources.metadata(parent_stack, policy)
+ return not metadata.get('scaling_in_progress')
+ return False