)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":37598,"name":"Ivan Anfimov","display_name":"Ivan Anfimov","email":"lazekteam@gmail.com","username":"anfimovir"},"change_message_id":"4f2182799c070d55a577ea297b958c75cc089993","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"46177fc7_68824c45","updated":"2026-09-25 02:50:27.000000000","message":"@dcu995@gmail.com I think, need rebase this MR after - https://review.opendev.org/c/openstack/trove/+/1006696","commit_id":"05f0056e3a9e1db6eff052481b534c94d51b4b3f"},{"author":{"_account_id":36080,"name":"Erkin Mussurmankulov","display_name":"Eric","email":"erkin.mussurmankulov@ps.kz","username":"mongoose404","status":"PS Cloud services employee"},"change_message_id":"cb19dedc597129bc0a967f4451ec97af07e90b25","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"d0788d3d_4f3535b2","in_reply_to":"2aac3027_8039d035","updated":"2026-09-25 17:28:05.000000000","message":"I think it\u0027s better to address both issues in two separate MRs, then.\n\nThe error I get is:\n\n```\nTraceback (most recent call last):                                                                             \n  File \"/usr/lib/python3.9/site-packages/trove/taskmanager/manager.py\", line 448, in create_instance           \n    self._create_instance(context, instance_id, name, flavor,                                                  \n  File \"/usr/lib/python3.9/site-packages/trove/taskmanager/manager.py\", line 399, in _create_instance          \n    self._create_replication_slave(context, instance_id, name,                                                 \n  File \"/usr/lib/python3.9/site-packages/trove/taskmanager/manager.py\", line 374, in _create_replication_slave \n    instance_tasks.create_instance(                                                                            \n  File \"/usr/lib/python3.9/site-packages/trove/taskmanager/models.py\", line 612, in create_instance            \n    cinder_snapshot_id \u003d backup.location.split(\"/\")[-1]                                                        \nAttributeError: \u0027NoneType\u0027 object has no attribute \u0027split\u0027                                                     \n```\n\nBut it appears at random. I cannot reproduce it intentionally.","commit_id":"05f0056e3a9e1db6eff052481b534c94d51b4b3f"},{"author":{"_account_id":37632,"name":"Dmitriy Chubinidze","email":"dcu995@gmail.com","username":"chubinidzedr"},"change_message_id":"c0b1c56d5c06296d657fa12fb1322d278727ac6d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"2aac3027_8039d035","in_reply_to":"2ca026ef_79606b27","updated":"2026-09-25 09:34:47.000000000","message":"Right, but there are a few different bugs appearing in this task, so it\u0027s probably worth figuring out which one you\u0027ve hit. First thing I saw was ``\u0027NoneType\u0027 object has no attribute \u0027split\u0027`` in ``taskmanager/models.py (backup.location.split(...))``. \nIf I understand correctly, the replica code assumed the Cinder snapshot record already had a location, but it was never written. Did you get any errors related to this?\n\nSecond one was ``ModelNotFoundError: DBInstance Not Found`` on conductor\u0027s side, that\u0027s the one encryption fixes. Taskmanager tags its rpc message with an instance id from ``CONF.guest_id``, which only exists on guest agents, not control-plane services, so it sends ``instance_id: None``, and conductor does ``DBInstance.find_by(id\u003dNone)`` looking for a decrypt key and rejects the message. So the update is rejected before it\u0027s ever applied, and master-replica sync doesn\u0027t have anything to do with it. With encryption on, this fails every time.","commit_id":"05f0056e3a9e1db6eff052481b534c94d51b4b3f"},{"author":{"_account_id":36080,"name":"Erkin Mussurmankulov","display_name":"Eric","email":"erkin.mussurmankulov@ps.kz","username":"mongoose404","status":"PS Cloud services employee"},"change_message_id":"3dc7db5d7700b77b96f48ab6dc47b6cd10e1d98f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"2ca026ef_79606b27","in_reply_to":"46177fc7_68824c45","updated":"2026-09-25 06:28:57.000000000","message":"Yes, right. The single responsibility principle is enforced 🙂\n\nI hit the bug that this MR addressed yesterday while running tests in our DEV OpenStack environment. But it\u0027s an intermittent bug and appears under certain conditions. I haven\u0027t yet completely understood exactly which ones.\n\nMy first thought was that it\u0027s a master-replica sync delay. In that case, this issue may not be fully resolved by this MR because the checking loop may fetch data from the master, and the instance creation process may read from a not-yet-synced replica.\n\n@dcu995@gmail.com have you unraveled the nature of this bug? Encryption seems kind of unrelated 🤔","commit_id":"05f0056e3a9e1db6eff052481b534c94d51b4b3f"},{"author":{"_account_id":36080,"name":"Erkin Mussurmankulov","display_name":"Eric","email":"erkin.mussurmankulov@ps.kz","username":"mongoose404","status":"PS Cloud services employee"},"change_message_id":"de0203e751d16454840aa30527282584c3d90a16","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"ab50068c_10dd3225","in_reply_to":"d0788d3d_4f3535b2","updated":"2026-09-25 17:30:48.000000000","message":"I will apply this patch and run tests to see whether this approach really works.","commit_id":"05f0056e3a9e1db6eff052481b534c94d51b4b3f"},{"author":{"_account_id":36080,"name":"Erkin Mussurmankulov","display_name":"Eric","email":"erkin.mussurmankulov@ps.kz","username":"mongoose404","status":"PS Cloud services employee"},"change_message_id":"5c055d333b1da915f6e8e5038f479c5d39f52cf5","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"51def72f_6b5bdc28","updated":"2026-09-30 07:53:12.000000000","message":"Well, after several days of testing without this patch, I wasn\u0027t able to see this error again 🤔\nIt seems very rare, but I have a feeling it\u0027s there.\nProbably we should intentionally delay the sync process between the replica and the primary to reproduce this error. I will ask our DevOps team if it\u0027s possible.","commit_id":"4f57c15eb51b5b677f22c45ee8425b833dd69a0c"}],"trove/taskmanager/models.py":[{"author":{"_account_id":39300,"name":"Olessya Khussainova","display_name":"Olessya Khussainova","email":"olessya.khussainova@ps.kz","username":"Cellofun"},"change_message_id":"a5dac0b784facf8a8dde9c4a8dde859999667eea","unresolved":true,"context_lines":[{"line_number":1405,"context_line":"        LOG.info(\"Initiating datastore restart on instance %s.\", self.id)"},{"line_number":1406,"context_line":"        try:"},{"line_number":1407,"context_line":"            self.guest.restart()"},{"line_number":1408,"context_line":"        except GuestTimeout:"},{"line_number":1409,"context_line":"            LOG.error(\"Timed out restarting datastore on instance %s.\","},{"line_number":1410,"context_line":"                      self.id)"},{"line_number":1411,"context_line":"            self._refresh_datastore_status()"}],"source_content_type":"text/x-python","patch_set":4,"id":"dee7ae24_83fba7a9","side":"PARENT","line":1408,"updated":"2026-09-25 07:13:06.000000000","message":"Please rebase this change. The current diff appears to revert part of [the previously merged fix](https://review.opendev.org/c/openstack/trove/+/1005508), removing the `GuestTimeout` handler and its tests.","commit_id":"cdd7324c9e274318ec576353f84be40fd4e0f9d6"},{"author":{"_account_id":37632,"name":"Dmitriy Chubinidze","email":"dcu995@gmail.com","username":"chubinidzedr"},"change_message_id":"e3db7d06c286cd4dbb1c556232d0f08da3657d0b","unresolved":false,"context_lines":[{"line_number":1405,"context_line":"        LOG.info(\"Initiating datastore restart on instance %s.\", self.id)"},{"line_number":1406,"context_line":"        try:"},{"line_number":1407,"context_line":"            self.guest.restart()"},{"line_number":1408,"context_line":"        except GuestTimeout:"},{"line_number":1409,"context_line":"            LOG.error(\"Timed out restarting datastore on instance %s.\","},{"line_number":1410,"context_line":"                      self.id)"},{"line_number":1411,"context_line":"            self._refresh_datastore_status()"}],"source_content_type":"text/x-python","patch_set":4,"id":"45d0ed9a_f831c622","side":"PARENT","line":1408,"in_reply_to":"dee7ae24_83fba7a9","updated":"2026-09-25 07:57:54.000000000","message":"Done","commit_id":"cdd7324c9e274318ec576353f84be40fd4e0f9d6"}]}