Cleanup some todo's for refresh()

There were some TODO tags in the code to switch to using object refresh()
calls once that was available. We have that now, so clean up and switch a
few instances to call refresh instead of fetching a new object.

Change-Id: I7d7c4f7d9f9cc94d9ef4e0d5d430513cb3581f16

cinder/backup/chunkeddriver.py

@@ -484,7 +484,7 @@ class ChunkedBackupDriver(driver.BackupDriver):
             # First of all, we check the status of this backup. If it
             # has been changed to delete or has been deleted, we cancel the
             # backup process to do forcing delete.
-            backup = objects.Backup.get_by_id(self.context, backup.id)
+            backup.refresh()
             if backup.status in (fields.BackupStatus.DELETING,
                                  fields.BackupStatus.DELETED):
                 is_backup_canceled = True

cinder/volume/manager.py

@@ -1380,7 +1380,7 @@ class VolumeManager(manager.CleanableManager,
 
         try:
             self.create_volume(ctx, image_volume, allow_reschedule=False)
-            image_volume = objects.Volume.get_by_id(ctx, image_volume.id)
+            image_volume.refresh()
             if image_volume.status != 'available':
                 raise exception.InvalidVolume(_('Volume is not available.'))
 
@@ -2028,8 +2028,8 @@ class VolumeManager(manager.CleanableManager,
         # Wait for new_volume to become ready
         starttime = time.time()
         deadline = starttime + CONF.migration_create_volume_timeout_secs
-        # TODO(thangp): Replace get_by_id with refresh when it is available
-        new_volume = objects.Volume.get_by_id(ctxt, new_volume.id)
+
+        new_volume.refresh()
         tries = 0
         while new_volume.status != 'available':
             tries += 1
@@ -2048,9 +2048,7 @@ class VolumeManager(manager.CleanableManager,
                 raise exception.VolumeMigrationFailed(reason=msg)
             else:
                 time.sleep(tries ** 2)
-            # TODO(thangp): Replace get_by_id with refresh when it is
-            # available
-            new_volume = objects.Volume.get_by_id(ctxt, new_volume.id)
+            new_volume.refresh()
 
         # Set skipped value to avoid calling
         # function except for _create_raw_volume
@@ -2927,8 +2925,7 @@ class VolumeManager(manager.CleanableManager,
             if group_snapshot:
                 try:
                     # Check if group_snapshot still exists
-                    group_snapshot = objects.GroupSnapshot.get_by_id(
-                        context, group_snapshot.id)
+                    group_snapshot.refresh()
                 except exception.GroupSnapshotNotFound:
                     LOG.error("Create group from snapshot-%(snap)s failed: "
                               "SnapshotNotFound.",
@@ -2954,8 +2951,7 @@ class VolumeManager(manager.CleanableManager,
 
             if source_group:
                 try:
-                    source_group = objects.Group.get_by_id(
-                        context, source_group.id)
+                    source_group.refresh()
                 except exception.GroupNotFound:
                     LOG.error("Create group "
                               "from source group-%(group)s failed: "
@@ -4780,7 +4776,7 @@ class VolumeManager(manager.CleanableManager,
 
         replication_targets = []
         try:
-            group = objects.Group.get_by_id(ctxt, group.id)
+            group.refresh()
             if self.configuration.replication_device:
                 if ctxt.is_admin:
                     for rep_dev in self.configuration.replication_device: