)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":597,"name":"Pete Zaitcev","email":"zaitcev@kotori.zaitcev.us","username":"zaitcev"},"change_message_id":"413d3726adf60b01503943d84dbff5d656eeb1df","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"e1cd3894_93917731","updated":"2023-04-27 02:44:59.000000000","message":"I\u0027m okay with it, but reading the old code it appears that it should work just fine. Yes, the values are incorrect, but there\u0027s no aliasing or overflows, the inputs are IDs, and the results of all these calculations only compare with themselves. The only bad result is a log message that points to a wrong config. Because of this, I\u0027m curious what prompted the fix. My reading of the code tells that it could not be an undue test failure.","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"55eeccda6986e5b7fb4b11bb42b7a986f55ef6a5","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"0545d4d2_69ebe758","updated":"2023-04-27 22:16:59.000000000","message":"This is all bad and stupid anyway -- should have a much better patch up soon.","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"a4137c51ff9fc7e6f635db6fccf3f587a7434336","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"1318cd5b_7debe394","updated":"2023-04-27 15:13:00.000000000","message":"looks like this is working; nice fix Tim!","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"060c239bb154b066515b21be3dbab45a8cf09f25","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"7ab265c8_55a4c687","updated":"2023-04-27 00:00:46.000000000","message":"recheck\n\nhttps://bugs.launchpad.net/swift/+bug/2017021\n\nOK, that particular oddity isn\u0027t the cause of the flakey test. (Not that I thought it was; we would sometimes run too *many* reconstructors before, not too few.)","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"f1c58c7561d7fb4c7e986a3027109af33fdae54a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"4c566527_ad370c4b","in_reply_to":"e1cd3894_93917731","updated":"2023-04-27 16:24:51.000000000","message":"\u003e I\u0027m curious what prompted the fix.\n\nJust a careful reading of the test while trying to run down https://bugs.launchpad.net/swift/+bug/2017021 -- though this doesn\u0027t fix it :-(","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":597,"name":"Pete Zaitcev","email":"zaitcev@kotori.zaitcev.us","username":"zaitcev"},"change_message_id":"1155ef331e77f390ae3a6f5cd79f8333ddc87be1","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"b4b38e5c_14039bfe","updated":"2023-04-28 03:44:39.000000000","message":"I didn\u0027t realize that config_number was appropriate here, it operates on a somewhat different principle.","commit_id":"88941ebe464139bcf5e8f30ba498dd42817dfe08"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"2f6792d106c0fc34c4713813ef4e2dec6deb85f7","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"ec0e7862_19e6d717","updated":"2023-04-28 10:23:42.000000000","message":"looking good, 3/3 successful checks","commit_id":"88941ebe464139bcf5e8f30ba498dd42817dfe08"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"b21ad4d3f7e9a5e62617202f8f87191d023b0542","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"4a61bf0b_82ee3f93","updated":"2023-04-28 00:36:42.000000000","message":"recheck\n\nI wanna see this guy pass a couple times to confirm the fix.","commit_id":"88941ebe464139bcf5e8f30ba498dd42817dfe08"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"b15b25b47d2d2b3e980217ecf33618153ebeca8a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"223965fb_691f27ea","updated":"2023-04-28 05:33:57.000000000","message":"recheck\n\nOne more time -- then I\u0027ll feel pretty good about this being the smoking gun.","commit_id":"88941ebe464139bcf5e8f30ba498dd42817dfe08"}],"test/probe/test_object_versioning.py":[{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"a4137c51ff9fc7e6f635db6fccf3f587a7434336","unresolved":true,"context_lines":[{"line_number":284,"context_line":"                break"},{"line_number":285,"context_line":"        else:"},{"line_number":286,"context_line":"            self.fail(\u0027unable to find fix_config in %r excluding %r \u0026 %r\u0027 % ("},{"line_number":287,"context_line":"                [(d[\u0027device\u0027], (d[\u0027id\u0027] % 4) + 1) for d in partner_nodes],"},{"line_number":288,"context_line":"                handoff_config, failed_config))"},{"line_number":289,"context_line":""},{"line_number":290,"context_line":"        self.reconstructor.once(number\u003dfix_config)"}],"source_content_type":"text/x-python","patch_set":1,"id":"811d2d4e_5075d3c5","line":287,"updated":"2023-04-27 15:13:00.000000000","message":"since I got this math wrong I wonder if we could move it to a helper - any node id should be able to convert directly to a config_num","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"a4137c51ff9fc7e6f635db6fccf3f587a7434336","unresolved":true,"context_lines":[{"line_number":287,"context_line":"                [(d[\u0027device\u0027], (d[\u0027id\u0027] % 4) + 1) for d in partner_nodes],"},{"line_number":288,"context_line":"                handoff_config, failed_config))"},{"line_number":289,"context_line":""},{"line_number":290,"context_line":"        self.reconstructor.once(number\u003dfix_config)"},{"line_number":291,"context_line":"        # validate object in all locations"},{"line_number":292,"context_line":"        missing \u003d []"},{"line_number":293,"context_line":"        etags \u003d set()"}],"source_content_type":"text/x-python","patch_set":1,"id":"f4713384_807e502e","line":290,"updated":"2023-04-27 15:13:00.000000000","message":"I\u0027m pretty sure getting the conf number wrong could have caused test failures","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"2f6792d106c0fc34c4713813ef4e2dec6deb85f7","unresolved":true,"context_lines":[{"line_number":287,"context_line":"                [(d[\u0027device\u0027], (d[\u0027id\u0027] % 4) + 1) for d in partner_nodes],"},{"line_number":288,"context_line":"                handoff_config, failed_config))"},{"line_number":289,"context_line":""},{"line_number":290,"context_line":"        self.reconstructor.once(number\u003dfix_config)"},{"line_number":291,"context_line":"        # validate object in all locations"},{"line_number":292,"context_line":"        missing \u003d []"},{"line_number":293,"context_line":"        etags \u003d set()"}],"source_content_type":"text/x-python","patch_set":1,"id":"a194d492_5d1152ed","line":290,"in_reply_to":"769d0d38_71ab73c6","updated":"2023-04-28 10:23:42.000000000","message":"nice debug @Tim!","commit_id":"d297095646515cec855604ce4bbf84a85f504059"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"f1c58c7561d7fb4c7e986a3027109af33fdae54a","unresolved":true,"context_lines":[{"line_number":287,"context_line":"                [(d[\u0027device\u0027], (d[\u0027id\u0027] % 4) + 1) for d in partner_nodes],"},{"line_number":288,"context_line":"                handoff_config, failed_config))"},{"line_number":289,"context_line":""},{"line_number":290,"context_line":"        self.reconstructor.once(number\u003dfix_config)"},{"line_number":291,"context_line":"        # validate object in all locations"},{"line_number":292,"context_line":"        missing \u003d []"},{"line_number":293,"context_line":"        etags \u003d set()"}],"source_content_type":"text/x-python","patch_set":1,"id":"769d0d38_71ab73c6","line":290,"in_reply_to":"f4713384_807e502e","updated":"2023-04-27 16:24:51.000000000","message":"The error was that we could pass 0 instead of 4 here -- which would cause us to run all reconstructors instead of just the one. See https://github.com/openstack/swift/blob/master/swift/common/manager.py#L576-L583\n\nBut, since we don\u0027t make assertions that there\u0027s still data on the handoff between here and processing the revert, it wouldn\u0027t cause any failures.","commit_id":"d297095646515cec855604ce4bbf84a85f504059"}]}