Add tenacity retries around ring sync check

This fixes a race when checking if object ring is synced.

(cherry picked from commit 0598ecf)

zaza/openstack/charm_tests/swift/tests.py

@@ -132,9 +132,15 @@ def test_904_set_weight_action_and_validate_rebalance(self):
         zaza.model.wait_for_agent_status()
         zaza.model.block_until_all_units_idle()
 
-        self.assertTrue(
-            swift_utils.is_ring_synced('swift-proxy', 'object',
-                                       self.TEST_EXPECTED_RING_HOSTS))
+        for attempt in tenacity.Retrying(
+                wait=tenacity.wait_fixed(2),
+                retry=tenacity.retry_if_exception_type(AssertionError),
+                reraise=True,
+                stop=tenacity.stop_after_attempt(10)):
+            with attempt:
+                self.assertTrue(
+                    swift_utils.is_ring_synced('swift-proxy', 'object',
+                                               self.TEST_EXPECTED_RING_HOSTS))
 
     def test_905_remove_device_action_and_validate_rebalance(self):
         """Remove device from object ring."""