)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"5c27373b941e3c9ce2193871df6ef41de5cd0e61","unresolved":true,"context_lines":[{"line_number":12,"context_line":"table having same cell name (eg. cell1)."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"This change adds unique constraint on \u0027name\u0027 field in"},{"line_number":15,"context_line":"CellMappings table in nova_api db."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Closes-Bug: #1923899"},{"line_number":18,"context_line":"Closes-Bug: #1736731"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"640d906a_ff9e3e55","line":15,"updated":"2023-03-09 09:16:22.000000000","message":"This is a change that restricts the DB schema so this might be a non backward compatible change on a system that has two cells with the same name. So we need to prepare for that. At least we need a pre-upgrade check that can detect such situation and warn the deployer that the db schema change will fail. And we need at least a release notes describing how to get out of the situation when you have two cells in the DB with the same name.","commit_id":"522a06c3d31aa2487ba2c97d32712b3958ad45d6"},{"author":{"_account_id":20733,"name":"Rajesh Tailor","email":"ratailor@redhat.com","username":"rajesht"},"change_message_id":"6b015c2dbdf2fa6abba91049b6c0917afbf4b730","unresolved":false,"context_lines":[{"line_number":12,"context_line":"table having same cell name (eg. cell1)."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"This change adds unique constraint on \u0027name\u0027 field in"},{"line_number":15,"context_line":"CellMappings table in nova_api db."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Closes-Bug: #1923899"},{"line_number":18,"context_line":"Closes-Bug: #1736731"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"c395121e_4af3c81f","line":15,"in_reply_to":"640d906a_ff9e3e55","updated":"2023-03-13 10:47:56.000000000","message":"Done","commit_id":"522a06c3d31aa2487ba2c97d32712b3958ad45d6"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3c2ddde0cb903c0706ae94d681b91353e74e1e47","unresolved":true,"context_lines":[{"line_number":9,"context_line":"As of now, if user tries to create a cell using templated"},{"line_number":10,"context_line":"urls and execute the same command (nova-manage cell_v2 create_cell)"},{"line_number":11,"context_line":"twice, then two different db entries are created in CellMappings"},{"line_number":12,"context_line":"table having same cell name (eg. cell1)."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"This change adds unique constraint on \u0027name\u0027 field in"},{"line_number":15,"context_line":"CellMappings table in nova_api db."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":4,"id":"299fda23_6d408d5e","line":12,"updated":"2023-04-17 12:15:04.000000000","message":"so the name there is just for humans so this shoudl not break anything\nif they have the same name.\n\nnova is going ot use the cell uuid for its internal usage.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"43f07180c4033c37b3e98d66b9100515de2cd168","unresolved":true,"context_lines":[{"line_number":15,"context_line":"CellMappings table in nova_api db."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Closes-Bug: #1923899"},{"line_number":18,"context_line":"Closes-Bug: #1736731"},{"line_number":19,"context_line":"Change-Id: If3210c0659ac0d81586731f948db7a3a899b0b3e"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":4,"id":"411cdebd_dfdf0277","line":18,"updated":"2023-04-17 14:09:18.000000000","message":"This is a dupe of the one above. No need to list both, IMHO.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"}],"/PATCHSET_LEVEL":[{"author":{"_account_id":20733,"name":"Rajesh Tailor","email":"ratailor@redhat.com","username":"rajesht"},"change_message_id":"6b015c2dbdf2fa6abba91049b6c0917afbf4b730","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"5e0c9498_f3a4d320","updated":"2023-03-13 10:47:56.000000000","message":"Thanks for review, I have added upgrade check.","commit_id":"1e385646a6512a17d5d292034761a9104998d220"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"49c0e3825f4c8566a7e098c677eefee716dcfd23","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"4815e4e3_2b81ec44","updated":"2023-04-17 12:17:50.000000000","message":"-1 because i think we need to split this into two patches to allow the nova-status change to be backported while the db change is not for SLURP releases.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"43f07180c4033c37b3e98d66b9100515de2cd168","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"4919056b_733e9bdf","updated":"2023-04-17 14:09:18.000000000","message":"I\u0027m a bit mixed on whether or not this is really worth the trouble, especially since we could handle it (albeit with less atomicity) with just a check-then-set approach in nova-manage. I think blocker migrations should be reserved for things where we\u0027re changing the structure of linked objects, fixing a field that needs to be a different type, or adding meaningful integrity constraints. This is admin tooling behavioral sugar, and a blocker migration to enable it seems like a large hammer.\n\nIf we\u0027re going to do it this way, I think we need warnings in nova-status (and perhaps logging at load time in the actual services) at least one SLURP release before we introduce a blocker migration like this.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3c2ddde0cb903c0706ae94d681b91353e74e1e47","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"aefd0260_e11e85ab","updated":"2023-04-17 12:15:04.000000000","message":"im not nessisaly againt this change but i dont think we should backport it upstream or downstream.\n\n+1 for now, i would like to see what others think.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":20733,"name":"Rajesh Tailor","email":"ratailor@redhat.com","username":"rajesht"},"change_message_id":"8612bf0fc6cc536b9f621058288242b56550553f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"3d8660d3_0be37c56","updated":"2023-03-15 11:11:05.000000000","message":"recheck openstack-tox-py38 test failed","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3cd5456008df30c89c434f87af408e6621bd1875","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"82cb86e3_f13ab5e2","in_reply_to":"4919056b_733e9bdf","updated":"2023-04-17 14:22:45.000000000","message":"i was wonderign about just doing this in nova-mangage too.\n\nthat would still break script that generated these dupe but i think that is less bad then breaking upgrades and or the cost of a blocker migration.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"}],"nova/cmd/status.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"49c0e3825f4c8566a7e098c677eefee716dcfd23","unresolved":true,"context_lines":[{"line_number":127,"context_line":"                    \u0027retry.\u0027)"},{"line_number":128,"context_line":"            return upgradecheck.Result(upgradecheck.Code.FAILURE, msg)"},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"        # check if duplicate cells with same name are present or not"},{"line_number":131,"context_line":"        cell_names_set \u003d set([mapping.name for mapping in cell_mappings])"},{"line_number":132,"context_line":"        if count !\u003d len(cell_names_set):"},{"line_number":133,"context_line":"            msg \u003d _(\u0027There are at least two cells with same name. \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"0cf39543_16e1df5a","line":130,"updated":"2023-04-17 12:17:50.000000000","message":"im wondering if this shoudl be done in a seperate change so that the nova-status check can be backported to antelope.\n\nthe schema change would then take effect in bobcat.\nfor SLURP release that would allow them to detect this on antelop and correct the issue before upgradeing to 2024.1","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"}],"nova/db/api/migrations/versions/88ef7b0ca805_add_uniq_cell_mappings_name.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3c2ddde0cb903c0706ae94d681b91353e74e1e47","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":4,"id":"c8e9064f_87043cd4","line":32,"updated":"2023-04-17 12:15:04.000000000","message":"this schema migration will obviously fail if the constraint is violated\nso i guess the nova status check is required to inform operators fo that before the upgrade.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"43f07180c4033c37b3e98d66b9100515de2cd168","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":4,"id":"2689a1d3_ab6ad93b","line":32,"in_reply_to":"c8e9064f_87043cd4","updated":"2023-04-17 14:09:18.000000000","message":"Right, and since the human-friendly name has been allowed to be duplicated up until now, you have to assume that some people will have such dupes in their database. If they never care about looking at the cell records themselves, it\u0027s certainly possible some script has added every cell as the same name.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"}],"nova/db/api/models.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3c2ddde0cb903c0706ae94d681b91353e74e1e47","unresolved":true,"context_lines":[{"line_number":137,"context_line":"    \"\"\"Contains information on communicating with a cell\"\"\""},{"line_number":138,"context_line":"    __tablename__ \u003d \u0027cell_mappings\u0027"},{"line_number":139,"context_line":"    __table_args__ \u003d ("},{"line_number":140,"context_line":"        sa.Index(\u0027uuid_idx\u0027, \u0027uuid\u0027),"},{"line_number":141,"context_line":"        schema.UniqueConstraint(\u0027uuid\u0027, name\u003d\u0027uniq_cell_mappings0uuid\u0027),"},{"line_number":142,"context_line":"        schema.UniqueConstraint(\u0027name\u0027, name\u003d\u0027uniq_cell_mappings0name\u0027),"},{"line_number":143,"context_line":"    )"}],"source_content_type":"text/x-python","patch_set":4,"id":"d616cbd5_7aeaf4c7","line":140,"range":{"start_line":140,"start_character":8,"end_line":140,"end_character":37},"updated":"2023-04-17 12:15:04.000000000","message":"this is technically not required by the way.\n\na UniqueConstraint creats an index.\nfixing that woudl be an unrealted change but we could do it now to minimes the db schema chagnes.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"}],"releasenotes/notes/fix-duplicate-cell-names-84427a36f7e38f7c.yaml":[{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"43f07180c4033c37b3e98d66b9100515de2cd168","unresolved":true,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"upgrade:"},{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    This is backword incompatible change and it restricts schema update"},{"line_number":5,"context_line":"    if there are duplicate cells with same name."},{"line_number":6,"context_line":"    To update duplicate cell, use command ``nova-manage cell_v2 list_cells``"},{"line_number":7,"context_line":"    to get list of cells and then use ``nova-manage cell_v2 update_cell``"}],"source_content_type":"text/x-yaml","patch_set":4,"id":"a98af09d_4bb1b34b","line":4,"range":{"start_line":4,"start_character":12,"end_line":4,"end_character":20},"updated":"2023-04-17 14:09:18.000000000","message":"\"a backwards\"","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"43f07180c4033c37b3e98d66b9100515de2cd168","unresolved":true,"context_lines":[{"line_number":5,"context_line":"    if there are duplicate cells with same name."},{"line_number":6,"context_line":"    To update duplicate cell, use command ``nova-manage cell_v2 list_cells``"},{"line_number":7,"context_line":"    to get list of cells and then use ``nova-manage cell_v2 update_cell``"},{"line_number":8,"context_line":"    command to update cell name."},{"line_number":9,"context_line":"fixes:"},{"line_number":10,"context_line":"  - |"},{"line_number":11,"context_line":"    This change fixes the issue where duplicate cell names can be created"}],"source_content_type":"text/x-yaml","patch_set":4,"id":"714c9931_b6984c0d","line":8,"updated":"2023-04-17 14:09:18.000000000","message":"I don\u0027t think this is a reasonable thing to throw into the migrations as a blocker with no warning, which I believe is Sean\u0027s point. We sometimes do need blocking migrations, but I think they need far better justification than this, and far more warning.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3cd5456008df30c89c434f87af408e6621bd1875","unresolved":true,"context_lines":[{"line_number":5,"context_line":"    if there are duplicate cells with same name."},{"line_number":6,"context_line":"    To update duplicate cell, use command ``nova-manage cell_v2 list_cells``"},{"line_number":7,"context_line":"    to get list of cells and then use ``nova-manage cell_v2 update_cell``"},{"line_number":8,"context_line":"    command to update cell name."},{"line_number":9,"context_line":"fixes:"},{"line_number":10,"context_line":"  - |"},{"line_number":11,"context_line":"    This change fixes the issue where duplicate cell names can be created"}],"source_content_type":"text/x-yaml","patch_set":4,"id":"7933dd26_b3e9751b","line":8,"in_reply_to":"714c9931_b6984c0d","updated":"2023-04-17 14:22:45.000000000","message":"yes that is my concern.\n\ni dont really think we cna just do this in one release without warning.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3c2ddde0cb903c0706ae94d681b91353e74e1e47","unresolved":true,"context_lines":[{"line_number":11,"context_line":"    This change fixes the issue where duplicate cell names can be created"},{"line_number":12,"context_line":"    for cell_mappings entries."},{"line_number":13,"context_line":"    This change adds a unique constraint on name field in cell_mappings"},{"line_number":14,"context_line":"    table in nova_api database."}],"source_content_type":"text/x-yaml","patch_set":4,"id":"9ec94f7f_e9ad188f","line":14,"updated":"2023-04-17 12:15:04.000000000","message":"given this should not break anything im not sure that fixes makes sense here.\n\nwe are decied to make the cell names unique but they were not used for anythign in nova jsut the uuids so the fact they could be duplicates is not really a bug.\n\nthe unqiuceness guarentee that we are intoducing feel more like a mini feature to me.\n\ni would just keep the upgrades section honestly and drop fixes.\n\nlets see what others think.","commit_id":"c928fbdcc10eddc0e4be605181c061d3e42de46e"}]}