)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"9682ff023f49a2e2a0aeefbf36e144ef1122fc2e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"7f05aaa2_bdf12677","updated":"2026-11-05 14:48:33.000000000","message":"Given the maintenance writes commen inline this might be more complicated than I anticipated. I\u0027m totally OK to delay this after CCLM and only land the smaller scope change in https://review.opendev.org/c/openstack/nova/+/1012083","commit_id":"352e04ece71568a39d59583e30768c569b44ccc1"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"61ad7a1a3d52a4275d2de607890570929ff6b14c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"bb9448c8_bf8bdd32","updated":"2026-11-04 19:25:27.000000000","message":"recheck ssh timeout in one test, unrelated to the allocations","commit_id":"352e04ece71568a39d59583e30768c569b44ccc1"}],"nova/scheduler/client/report.py":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"9682ff023f49a2e2a0aeefbf36e144ef1122fc2e","unresolved":true,"context_lines":[{"line_number":1710,"context_line":"        payload[\u0027project_id\u0027] \u003d project_id"},{"line_number":1711,"context_line":"        payload[\u0027user_id\u0027] \u003d user_id"},{"line_number":1712,"context_line":""},{"line_number":1713,"context_line":"        # Instance claims always use at least 1.38, even when the allocation"},{"line_number":1714,"context_line":"        # candidate was requested with an older microversion."},{"line_number":1715,"context_line":"        payload[\u0027consumer_generation\u0027] \u003d consumer_generation"},{"line_number":1716,"context_line":"        payload[\u0027consumer_type\u0027] \u003d CONSUMER_TYPE_INSTANCE"}],"source_content_type":"text/x-python","patch_set":2,"id":"0ae2ea4e_d368dae1","line":1713,"updated":"2026-11-05 14:48:33.000000000","message":"hm, I thought we kept the two version in sync as we are directly re-using the candidate json blob as the allocation blob in the request. So if the version is too far then it might not be compatible to do so. I guess this is safe to do now that the placement API become pretty stable. I guess a comment in the allocation candidate request microversion would help requesting to keep the a_c and allocation version sync or at least compatible.","commit_id":"352e04ece71568a39d59583e30768c569b44ccc1"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"9682ff023f49a2e2a0aeefbf36e144ef1122fc2e","unresolved":true,"context_lines":[{"line_number":2037,"context_line":"            version\u003dCONSUMER_GENERATION_VERSION):"},{"line_number":2038,"context_line":"        # Only callers that know the consumer\u0027s role supply a type. Generic"},{"line_number":2039,"context_line":"        # maintenance writes can also touch migration consumers and must not"},{"line_number":2040,"context_line":"        # assume they are instances. Preserve any newer requested version."},{"line_number":2041,"context_line":"        if \u0027consumer_type\u0027 in payload:"},{"line_number":2042,"context_line":"            version \u003d max(version, CONSUMER_TYPE_VERSION,"},{"line_number":2043,"context_line":"                          key\u003dversionutils.convert_version_to_tuple)"}],"source_content_type":"text/x-python","patch_set":2,"id":"f1cad276_9fc2b65e","line":2040,"updated":"2026-11-05 14:48:33.000000000","message":"I\u0027m a bit hesitant about this. Are those generic maintenance writes will create new allocations or just reading them and putting them back? If the former then after this change nova will still create new allocations without type specified. If the latter then we can bump the read microversion and the default write microversion to avoid loosing the type during a read/write maintenance.","commit_id":"352e04ece71568a39d59583e30768c569b44ccc1"}]}