)]}'
{"nova/network/neutronv2/api.py":[{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"9ff849526d26617769495a90dd0afa2c59b599bb","unresolved":false,"context_lines":[{"line_number":3287,"context_line":"        # shelve scenario). If _unbind_ports was going to be leveraged here, it"},{"line_number":3288,"context_line":"        # would have to be adjusted a bit since it currently clears the"},{"line_number":3289,"context_line":"        # device_id field on the port which is not what we\u0027d want for shelve."},{"line_number":3290,"context_line":"        # TODO(brizhang): The server creation failure should be handled."},{"line_number":3291,"context_line":"        # When rescheduling is performed, the network port is cleaned up,"},{"line_number":3292,"context_line":"        # otherwise there will be network port residual. It may cause the"},{"line_number":3293,"context_line":"        # server to generate two IPs on a network after reschedule"}],"source_content_type":"text/x-python","patch_set":2,"id":"3fa7e38b_fb4b6cad","line":3290,"updated":"2019-11-04 00:56:02.000000000","message":"This is already handled on reschedule for neutron with a different method:\n\nhttps://github.com/openstack/nova/blob/b377f1c0cdeb13cad2b2549b967dd00ffa195011/nova/compute/manager.py#L2200\n\nand here:\n\nhttps://github.com/openstack/nova/blob/b377f1c0cdeb13cad2b2549b967dd00ffa195011/nova/conductor/manager.py#L687","commit_id":"44346c133e776be7a444db52d06a19ca7208f63f"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"b99d1798aef8a0bd3a4fba8dd9d8ed34b2dd04f8","unresolved":false,"context_lines":[{"line_number":3287,"context_line":"        # shelve scenario). If _unbind_ports was going to be leveraged here, it"},{"line_number":3288,"context_line":"        # would have to be adjusted a bit since it currently clears the"},{"line_number":3289,"context_line":"        # device_id field on the port which is not what we\u0027d want for shelve."},{"line_number":3290,"context_line":"        # TODO(brizhang): The server creation failure should be handled."},{"line_number":3291,"context_line":"        # When rescheduling is performed, the network port is cleaned up,"},{"line_number":3292,"context_line":"        # otherwise there will be network port residual. It may cause the"},{"line_number":3293,"context_line":"        # server to generate two IPs on a network after reschedule"}],"source_content_type":"text/x-python","patch_set":2,"id":"3fa7e38b_2ddda040","line":3290,"in_reply_to":"3fa7e38b_65631ddf","updated":"2019-11-18 19:05:52.000000000","message":"\u003e If so, I think we should use \"self._cleanup_allocated_networks(context,\n \u003e instance, requested_networks))\" instead of use \"self.network_api.cleanup_instance_network_on_host(context,\n \u003e instance, self.host)\" in https://github.com/openstack/nova/blob/b377f1c0cdeb13cad2b2549b967dd00ffa195011/nova/compute/manager.py#L2207\n\nWe won\u0027t get to that line if you\u0027re using neutron:\n\nhttps://github.com/openstack/nova/blob/b377f1c0cdeb13cad2b2549b967dd00ffa195011/nova/compute/manager.py#L2198\n\nIs this TODO/fix just based purely on code inspection because as I\u0027ve said I don\u0027t think this is a bug and it\u0027s already handled if you\u0027re using neutron.","commit_id":"44346c133e776be7a444db52d06a19ca7208f63f"},{"author":{"_account_id":26458,"name":"Brin Zhang","email":"zhangbailin@inspur.com","username":"zhangbailin"},"change_message_id":"aaca47fcab077b1c5057afd5e25b1e28b07b5a68","unresolved":false,"context_lines":[{"line_number":3287,"context_line":"        # shelve scenario). If _unbind_ports was going to be leveraged here, it"},{"line_number":3288,"context_line":"        # would have to be adjusted a bit since it currently clears the"},{"line_number":3289,"context_line":"        # device_id field on the port which is not what we\u0027d want for shelve."},{"line_number":3290,"context_line":"        # TODO(brizhang): The server creation failure should be handled."},{"line_number":3291,"context_line":"        # When rescheduling is performed, the network port is cleaned up,"},{"line_number":3292,"context_line":"        # otherwise there will be network port residual. It may cause the"},{"line_number":3293,"context_line":"        # server to generate two IPs on a network after reschedule"}],"source_content_type":"text/x-python","patch_set":2,"id":"3fa7e38b_65631ddf","line":3290,"in_reply_to":"3fa7e38b_fb4b6cad","updated":"2019-11-08 07:02:13.000000000","message":"If so, I think we should use \"self._cleanup_allocated_networks(context, instance, requested_networks))\" instead of use \"self.network_api.cleanup_instance_network_on_host(context, instance, self.host)\" in https://github.com/openstack/nova/blob/b377f1c0cdeb13cad2b2549b967dd00ffa195011/nova/compute/manager.py#L2207","commit_id":"44346c133e776be7a444db52d06a19ca7208f63f"}]}