)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":2750,"name":"Sean Dague","email":"sean@dague.net","username":"sdague"},"change_message_id":"052e0fd9589c24f3d7d1fd24e8733df8566ca568","unresolved":false,"context_lines":[{"line_number":10,"context_line":"explicitly. Now _update_resource_tracker is called just from"},{"line_number":11,"context_line":"_complete_deletion so that both places that eventually call"},{"line_number":12,"context_line":"_complete_deletion will get _update_resource_tracker just once."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"Change-Id: I586c0305ea0e20d98697bc9173ca0f89dc2b228c"},{"line_number":15,"context_line":"Closes-Bug: #1227925"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":3,"id":"3afc51ec_e39a01c5","line":13,"updated":"2016-03-11 13:08:48.000000000","message":"In future I\u0027d change the commit message to something like this:\n\nWhen compute manager was started, and processed deleted instances, the resource_tracker was not getting updated. This\nmoves resource_tracer updates into common code.\n\nIt does mean that instance.delete() throwing an exception\nwill skip calling the resource_tracker update, but as \ndiscussed in the bug, at this point in the code things are\nall wibbly wobbly anyway, so not updating the resource \ntracker in those cases is not a big loss, as it would all\nbe pretty inconsistent data.","commit_id":"df31a4ddf391d0a2d1d47f83ea6a83ee3744ee37"},{"author":{"_account_id":5754,"name":"Alex Xu","email":"hejie.xu@intel.com","username":"xuhj"},"change_message_id":"7781168611baeb4c58d62cc39bf298ffdee2002c","unresolved":false,"context_lines":[{"line_number":10,"context_line":"explicitly. Now _update_resource_tracker is called just from"},{"line_number":11,"context_line":"_complete_deletion so that both places that eventually call"},{"line_number":12,"context_line":"_complete_deletion will get _update_resource_tracker just once."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"Change-Id: I586c0305ea0e20d98697bc9173ca0f89dc2b228c"},{"line_number":15,"context_line":"Closes-Bug: #1227925"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":3,"id":"3afc51ec_c34ee595","line":13,"in_reply_to":"3afc51ec_e39a01c5","updated":"2016-03-11 13:25:32.000000000","message":"If this bug is for updating deleted instance in RT when compute manager started, then looks like this fix is wrong. The deleted instance cleanup was done at ComputeManage.init_host, it is called before the \u0027pre_start_hook\u0027, in \u0027pre_start_hook\u0027, the RT will be initialized. So update the instance resource usage before RT initialization looks like useless.","commit_id":"df31a4ddf391d0a2d1d47f83ea6a83ee3744ee37"}],"nova/compute/manager.py":[{"author":{"_account_id":2750,"name":"Sean Dague","email":"sean@dague.net","username":"sdague"},"change_message_id":"a802546a3b7df96d83b86328327ad40f7ee259e1","unresolved":false,"context_lines":[{"line_number":903,"context_line":"                                                               instance)"},{"line_number":904,"context_line":"        quotas.reserve(project_id\u003dproject_id, user_id\u003duser_id, instances\u003d-1,"},{"line_number":905,"context_line":"                       cores\u003d-instance.vcpus, ram\u003d-instance.memory_mb)"},{"line_number":906,"context_line":"        self._complete_deletion(context,"},{"line_number":907,"context_line":"                                instance,"},{"line_number":908,"context_line":"                                bdms,"},{"line_number":909,"context_line":"                                quotas,"}],"source_content_type":"text/x-python","patch_set":3,"id":"3afc51ec_119f7604","line":906,"range":{"start_line":906,"start_character":8,"end_line":906,"end_character":31},"updated":"2016-03-11 13:00:57.000000000","message":"This is the other place it is called. This is called via _init_instance during compute manager restart.","commit_id":"df31a4ddf391d0a2d1d47f83ea6a83ee3744ee37"}],"nova/tests/unit/compute/test_compute.py":[{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"ca7c19d2b868fa51342e981a5d28c3a368919ac2","unresolved":false,"context_lines":[{"line_number":6829,"context_line":"        instance \u003d self._create_fake_instance_obj()"},{"line_number":6830,"context_line":"        instance.vm_state \u003d vm_states.DELETED"},{"line_number":6831,"context_line":"        self.compute._init_instance(admin_context, instance)"},{"line_number":6832,"context_line":"        update_tracker.assert_called_once_with(admin_context, instance)"},{"line_number":6833,"context_line":""},{"line_number":6834,"context_line":"    def test_init_instance_for_partial_deletion(self):"},{"line_number":6835,"context_line":"        admin_context \u003d context.get_admin_context()"}],"source_content_type":"text/x-python","patch_set":1,"id":"5aef4532_de8978cc","line":6832,"updated":"2016-03-07 20:53:11.000000000","message":"I tried to come up with less lame tests that actually measured the impact on the resource tracker but these tests are completely opaque to me, so extracting the tricks to get anything to happen passed me by. Advice welcome.","commit_id":"6b51ee965d0afc2c5e95d9db99c1cf2702fde8ff"}]}