)]}'
{"nova/compute/manager.py":[{"author":{"_account_id":10135,"name":"Lee Yarwood","display_name":"Lee Yarwood","email":"lyarwood@redhat.com","username":"lyarwood"},"change_message_id":"7bd15de853bbc42434b33578b7526f03bffb7403","unresolved":false,"context_lines":[{"line_number":2733,"context_line":"            bdms \u003d list(block_device.get_bdms_to_connect("},{"line_number":2734,"context_line":"                bdms, exclude_root_mapping\u003dTrue))"},{"line_number":2735,"context_line":"            bdms.insert(0, new_root_bdm)"},{"line_number":2736,"context_line":""},{"line_number":2737,"context_line":"        new_block_device_info \u003d attach_block_devices(context, instance, bdms)"},{"line_number":2738,"context_line":""},{"line_number":2739,"context_line":"        instance.task_state \u003d task_states.REBUILD_SPAWNING"}],"source_content_type":"text/x-python","patch_set":9,"id":"1a122d0e_2122e001","line":2736,"updated":"2016-04-18 14:39:45.000000000","message":"Do we really want to recreate the root volume in all cases? Wouldn\u0027t that be unexpected during an evacuation as the user would assume that the volume would persist away from either compute node?","commit_id":"b35a7ac590a7f14b5a9e91d3c7d150d354f5e78b"},{"author":{"_account_id":20873,"name":"Laszlo Hegedus","email":"hegedues.laszlo@gmail.com","username":"laszloh"},"change_message_id":"86d1a1f5c337f286334ccf9042680efff1dc882a","unresolved":false,"context_lines":[{"line_number":2733,"context_line":"            bdms \u003d list(block_device.get_bdms_to_connect("},{"line_number":2734,"context_line":"                bdms, exclude_root_mapping\u003dTrue))"},{"line_number":2735,"context_line":"            bdms.insert(0, new_root_bdm)"},{"line_number":2736,"context_line":""},{"line_number":2737,"context_line":"        new_block_device_info \u003d attach_block_devices(context, instance, bdms)"},{"line_number":2738,"context_line":""},{"line_number":2739,"context_line":"        instance.task_state \u003d task_states.REBUILD_SPAWNING"}],"source_content_type":"text/x-python","patch_set":9,"id":"1a122d0e_cf432041","line":2736,"in_reply_to":"1a122d0e_2122e001","updated":"2016-04-19 13:49:06.000000000","message":"You are right. I moved the code into the block where recreate is False (i.e., no evacuation).\n\nWhat do you think of the case when the image id in the volume is the same as the one that is used for rebuild? Should we still recreate the volume or should we leave it as it was? For now, I would stick to leaving it untouched in this case.\n\nAdditionally, I added the removal of the volume in case delete_on_termination is True.","commit_id":"b35a7ac590a7f14b5a9e91d3c7d150d354f5e78b"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"54d8a1eb3912a576c65f979e16e30ec584bf76da","unresolved":false,"context_lines":[{"line_number":2733,"context_line":"            bdms \u003d list(block_device.get_bdms_to_connect("},{"line_number":2734,"context_line":"                bdms, exclude_root_mapping\u003dTrue))"},{"line_number":2735,"context_line":"            bdms.insert(0, new_root_bdm)"},{"line_number":2736,"context_line":""},{"line_number":2737,"context_line":"        new_block_device_info \u003d attach_block_devices(context, instance, bdms)"},{"line_number":2738,"context_line":""},{"line_number":2739,"context_line":"        instance.task_state \u003d task_states.REBUILD_SPAWNING"}],"source_content_type":"text/x-python","patch_set":9,"id":"bab6814e_f966e027","line":2736,"in_reply_to":"1a122d0e_cf432041","updated":"2016-05-19 08:40:57.000000000","message":"I think during a rebuild it is valid to rebuild with the same image. I mean I have an instance booted from an image based volume but there is some disk corruption and I want to re-image the instance with the same image.","commit_id":"b35a7ac590a7f14b5a9e91d3c7d150d354f5e78b"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"0392d00595a8779dd70c29b4b76b0b730ab4dca5","unresolved":false,"context_lines":[{"line_number":2556,"context_line":"                                block_device_info\u003dblock_device_info)"},{"line_number":2557,"context_line":"            root_bdm \u003d block_device.get_root_bdm(bdms)"},{"line_number":2558,"context_line":"            if (root_bdm and root_bdm.is_volume and"},{"line_number":2559,"context_line":"                    not getattr(root_bdm, \u0027image_id\u0027, None) \u003d\u003d image_meta.id):"},{"line_number":2560,"context_line":"                old_vol_id \u003d root_bdm.volume_id"},{"line_number":2561,"context_line":"                new_root_bdm \u003d self._recreate_volume_based_bdm_from_image("},{"line_number":2562,"context_line":"                    context, image_meta, root_bdm)"}],"source_content_type":"text/x-python","patch_set":16,"id":"bab6814e_b9545821","line":2559,"range":{"start_line":2559,"start_character":24,"end_line":2559,"end_character":76},"updated":"2016-05-19 08:42:30.000000000","message":"I think re-imaging with the same image during rebuild is a valid use case, e.g fixing a corrupted root volume of an instance.","commit_id":"1215fc3a7053a51974e0687e2ff538ce8893dc40"},{"author":{"_account_id":20873,"name":"Laszlo Hegedus","email":"hegedues.laszlo@gmail.com","username":"laszloh"},"change_message_id":"3c79525638cced14d548ad293f59e4b8a4e6658d","unresolved":false,"context_lines":[{"line_number":2556,"context_line":"                                block_device_info\u003dblock_device_info)"},{"line_number":2557,"context_line":"            root_bdm \u003d block_device.get_root_bdm(bdms)"},{"line_number":2558,"context_line":"            if (root_bdm and root_bdm.is_volume and"},{"line_number":2559,"context_line":"                    not getattr(root_bdm, \u0027image_id\u0027, None) \u003d\u003d image_meta.id):"},{"line_number":2560,"context_line":"                old_vol_id \u003d root_bdm.volume_id"},{"line_number":2561,"context_line":"                new_root_bdm \u003d self._recreate_volume_based_bdm_from_image("},{"line_number":2562,"context_line":"                    context, image_meta, root_bdm)"}],"source_content_type":"text/x-python","patch_set":16,"id":"bab6814e_5696993f","line":2559,"range":{"start_line":2559,"start_character":24,"end_line":2559,"end_character":76},"in_reply_to":"bab6814e_b9545821","updated":"2016-05-20 07:12:39.000000000","message":"Good point. It\u0027s a valid use case.\n\nI\u0027m a bit concerned about recreating the root volume in all cases, though. If the bdm is not marked with delete_on_termination, then the newly created volumes could pile up.","commit_id":"1215fc3a7053a51974e0687e2ff538ce8893dc40"},{"author":{"_account_id":7730,"name":"Sahid Orentino Ferdjaoui","email":"sahid.ferdjaoui@industrialdiscipline.com","username":"sahid"},"change_message_id":"22cd5710904be75a623654890a400540d1f3883f","unresolved":false,"context_lines":[{"line_number":2529,"context_line":"                                            None,"},{"line_number":2530,"context_line":"                                            image_id\u003dimage_meta.id)"},{"line_number":2531,"context_line":"        self._await_block_device_map_created(context, new_volume[\u0027id\u0027])"},{"line_number":2532,"context_line":"        LOG.debug(\u0027Created new root volume. Updating bdm.\u0027)"},{"line_number":2533,"context_line":"        bdm.volume_id \u003d new_volume[\u0027id\u0027]"},{"line_number":2534,"context_line":"        bdm.connection_info \u003d None"},{"line_number":2535,"context_line":"        bdm.image_id \u003d image_meta.id"}],"source_content_type":"text/x-python","patch_set":18,"id":"9abb7d3a_d7428267","line":2532,"updated":"2016-05-31 13:55:44.000000000","message":"Can you refer the volume id.","commit_id":"47e862b0a4236bdea7ff76449aa764d5c79313c0"},{"author":{"_account_id":20873,"name":"Laszlo Hegedus","email":"hegedues.laszlo@gmail.com","username":"laszloh"},"change_message_id":"66e0f8295f186f8c057cb5ebb767b32cd792f59a","unresolved":false,"context_lines":[{"line_number":2529,"context_line":"                                            None,"},{"line_number":2530,"context_line":"                                            image_id\u003dimage_meta.id)"},{"line_number":2531,"context_line":"        self._await_block_device_map_created(context, new_volume[\u0027id\u0027])"},{"line_number":2532,"context_line":"        LOG.debug(\u0027Created new root volume. Updating bdm.\u0027)"},{"line_number":2533,"context_line":"        bdm.volume_id \u003d new_volume[\u0027id\u0027]"},{"line_number":2534,"context_line":"        bdm.connection_info \u003d None"},{"line_number":2535,"context_line":"        bdm.image_id \u003d image_meta.id"}],"source_content_type":"text/x-python","patch_set":18,"id":"9abb7d3a_6928df33","line":2532,"in_reply_to":"9abb7d3a_d7428267","updated":"2016-06-01 13:33:29.000000000","message":"Good idea. I\u0027ll add it in the next patch set.","commit_id":"47e862b0a4236bdea7ff76449aa764d5c79313c0"},{"author":{"_account_id":7730,"name":"Sahid Orentino Ferdjaoui","email":"sahid.ferdjaoui@industrialdiscipline.com","username":"sahid"},"change_message_id":"22cd5710904be75a623654890a400540d1f3883f","unresolved":false,"context_lines":[{"line_number":2559,"context_line":"                old_vol_id \u003d root_bdm.volume_id"},{"line_number":2560,"context_line":"                new_root_bdm \u003d self._recreate_volume_based_bdm_from_image("},{"line_number":2561,"context_line":"                    context, image_meta, root_bdm)"},{"line_number":2562,"context_line":"                if root_bdm.delete_on_termination:"},{"line_number":2563,"context_line":"                    try:"},{"line_number":2564,"context_line":"                        self.volume_api.delete(context, old_vol_id)"},{"line_number":2565,"context_line":"                    except Exception as exc:"}],"source_content_type":"text/x-python","patch_set":18,"id":"9abb7d3a_9a977feb","line":2562,"updated":"2016-05-31 13:55:44.000000000","message":"Hum.. At this point, don\u0027t we want to delete this volume in all case ?","commit_id":"47e862b0a4236bdea7ff76449aa764d5c79313c0"},{"author":{"_account_id":20873,"name":"Laszlo Hegedus","email":"hegedues.laszlo@gmail.com","username":"laszloh"},"change_message_id":"66e0f8295f186f8c057cb5ebb767b32cd792f59a","unresolved":false,"context_lines":[{"line_number":2559,"context_line":"                old_vol_id \u003d root_bdm.volume_id"},{"line_number":2560,"context_line":"                new_root_bdm \u003d self._recreate_volume_based_bdm_from_image("},{"line_number":2561,"context_line":"                    context, image_meta, root_bdm)"},{"line_number":2562,"context_line":"                if root_bdm.delete_on_termination:"},{"line_number":2563,"context_line":"                    try:"},{"line_number":2564,"context_line":"                        self.volume_api.delete(context, old_vol_id)"},{"line_number":2565,"context_line":"                    except Exception as exc:"}],"source_content_type":"text/x-python","patch_set":18,"id":"9abb7d3a_899ee304","line":2562,"in_reply_to":"9abb7d3a_9a977feb","updated":"2016-06-01 13:33:29.000000000","message":"I thought about this. I think it can happen that someone wants to boot another vm from the original volume. So, in case it is not marked delete_on_termination, I assume that the user wants to handle the deletion (and creation) of that volume.","commit_id":"47e862b0a4236bdea7ff76449aa764d5c79313c0"},{"author":{"_account_id":6849,"name":"Roman Podoliaka","email":"roman.podoliaka@gmail.com","username":"rpodolyaka"},"change_message_id":"b09a0f223bd1300285bc6a7897b2cc7d7475e7f6","unresolved":false,"context_lines":[{"line_number":2580,"context_line":""},{"line_number":2581,"context_line":"    def _recreate_volume_based_bdm_from_image(self, context, image_meta, bdm):"},{"line_number":2582,"context_line":"        new_volume \u003d self.volume_api.create(context,"},{"line_number":2583,"context_line":"                                            bdm.volume_size,"},{"line_number":2584,"context_line":"                                            bdm.device_name,"},{"line_number":2585,"context_line":"                                            None,"},{"line_number":2586,"context_line":"                                            image_id\u003dimage_meta.id)"}],"source_content_type":"text/x-python","patch_set":22,"id":"9ad45d7e_e7e7a2aa","line":2583,"updated":"2016-08-08 09:23:29.000000000","message":"What if the size of the new image is larger than the size of the current volume? Will creation of the new volume and, thus, rebuild fail?\n\nIMO, it\u0027s not really consistent with rebuilding of image-backed instances as is, where the only limit would be the root_gb value of the flavor used.","commit_id":"83d7a130b04a99c2240ba652c689021dd2762ac2"}]}