)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"f690e21435c94c0f6a3422765cb510ebda60fbac","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"9c9d168f_dedd38cb","updated":"2026-03-02 16:11:09.000000000","message":"this is the api contract to it was an input validation error before","commit_id":"d94ee9b26c9a36d48f8f4f5e68b85dcdd319d07e"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"d3c9d11aa1e2173ec78118f584a6b86b9d3b5ab3","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":4,"id":"a272d931_9558b668","in_reply_to":"9c9d168f_dedd38cb","updated":"2026-03-02 16:13:22.000000000","message":"No it has never returned validation error, until response schema validation was recently implemented.\nThis is proven by the fact that this test has been there and succeeding for some time.","commit_id":"d94ee9b26c9a36d48f8f4f5e68b85dcdd319d07e"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"45b5d81ccbd19a36d3e991ab34f06a8f08276f0c","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":4,"id":"058232cc_aeca0dd5","in_reply_to":"a272d931_9558b668","updated":"2026-03-02 17:07:52.000000000","message":"correct but it was documented as the only supproted value.","commit_id":"d94ee9b26c9a36d48f8f4f5e68b85dcdd319d07e"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"363316cf2d7c6f3162a46fa7d55f238ab8515e4f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"422fdda1_84fde9e5","updated":"2026-03-05 17:06:22.000000000","message":"I believe this is the correct fix. Using an older deployment, I am able to update and inspect quota class sets with different names. Those quota class sets may be useless (I don\u0027t believe we can do anything with them) but changing that would require an API microversion and should be blocked on the request schema side, not the response schema side.\n\n```\n❯ nova quota-class-update --instances 22 foo\nnova CLI is deprecated and will be removed in a future release\n\n❯ nova quota-class-show foo\nnova CLI is deprecated and will be removed in a future release\n+----------------------+-------+\n| Quota                | Limit |\n+----------------------+-------+\n| instances            | 22    |\n| cores                | 20    |\n| ram                  | 51200 |\n| metadata_items       | 128   |\n| key_pairs            | 100   |\n| server_groups        | 10    |\n| server_group_members | 10    |\n+----------------------+-------+\n\n❯ nova quota-class-update --instances 1 foo\nnova CLI is deprecated and will be removed in a future release\n\n❯ nova quota-class-show foo\nnova CLI is deprecated and will be removed in a future release\n+----------------------+-------+\n| Quota                | Limit |\n+----------------------+-------+\n| instances            | 1     |\n| cores                | 20    |\n| ram                  | 51200 |\n| metadata_items       | 128   |\n| key_pairs            | 100   |\n| server_groups        | 10    |\n| server_group_members | 10    |\n+----------------------+-------+\n```","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"f7560adfcbcf205f100b777e1d2ea1ab41df149e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"6ba93b6c_a0707c41","updated":"2026-03-02 16:29:25.000000000","message":"Thanks Takashi for fix. After re-checking the code, it sems error is fine but it is raised from wrong place. If nova does not support other class than \u0027default\u0027 (as per API contract [1]) then it should be validated before processing request. Current error is when code put the requested \u0027id\u0027 in response and response schema fail.\n\nI still think we should fix the test to solve this bug and this change for better error handling. Which might need discussion/checking for how non \u0027default\u0027 error in actual env and fail early accordingly. But i am good to merge this and have error handling as a separate discussion. I will change my vote to +2 once test is fixed.\n\nhttps://github.com/openstack/nova/blob/7a6e13454dc641ec592c17fb1b97f2f89d61377c/nova/api/openstack/compute/quota_classes.py#L62\n\n[1] https://docs.openstack.org/api-ref/compute/#quota-class-sets-os-quota-class-sets","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"c77489a2794e4abb3e0d379e59dcd618a0aa6acb","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"2f1f2bd6_6015337b","updated":"2026-03-05 18:59:19.000000000","message":"let\u0027s merge this to fix the behaviour change and how we can improve this api is separate step.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"4bea79758463e0da63402e9d64e035cc57a6978c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"730cf9e2_df04d635","updated":"2026-03-06 22:04:14.000000000","message":"recheck","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"8bf7cb86cbb84ea1de819301a9ab8e2908945a29","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"8a90cf1e_157b8fd1","updated":"2026-03-05 06:04:38.000000000","message":"recheck","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"010c3709d31d2e597a260fd00accdd68eeb346d0","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"a546a40e_9dafce84","updated":"2026-03-11 14:33:58.000000000","message":"recheck a different job failed this time.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"dc797c0006ffd2d800f80f78caa51a33978e0ddb","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"ad5a2a00_6fe30988","updated":"2026-03-11 16:56:38.000000000","message":"recheck another job (nova-alt-configurations) failed. Crossing finger this time.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"023a05a4861d11546aac06012d5114c8d12b4972","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"5f04af1f_fa97e912","updated":"2026-03-06 15:46:55.000000000","message":"recheck bad tox version broke several jobs.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"def08a3a50fd517dd7d18d3d90f976348f6f76ba","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"690d32a3_3becf7d3","updated":"2026-03-05 16:06:52.000000000","message":"recheck multi-cell failure is irrelevant","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"fa4bdf907da1f852a3ad98b1fdd197e2814d3f2f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"060a3a7a_166d4a9c","updated":"2026-03-06 02:39:56.000000000","message":"recheck nova-next failure","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"491085b6a19fd66c5ba304ec2774e7517d8c4fa1","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"0bed2ade_3ec50445","updated":"2026-03-07 07:33:32.000000000","message":"recheck nova-next failure is unrelated","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"d0864e1a9fa741c0a8376018293482091f91185d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"e2500148_19e11d85","updated":"2026-03-09 16:04:42.000000000","message":"recheck nova-next job failure is irrelevant","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"35c3ef8fd89872190bc15e2587c0de7ab6762ec4","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"c7767f2a_bfdb803f","in_reply_to":"3dc5717c_d5150a15","updated":"2026-03-05 17:10:44.000000000","message":"\u003e I still think we should fix the test to solve this bug and this change for better error handling. Which might need discussion/checking for how non \u0027default\u0027 error in actual env and fail early accordingly. But i am good to merge this and have error handling as a separate discussion.\n\nI don\u0027t believe we can change that test since setting the `default` quota class set has a knock on impact for other quotas, per [the node in the api-ref](https://docs.openstack.org/api-ref/compute/#quota-class-sets-os-quota-class-sets). The best we can do is remove the test but, as we\u0027ve seen, it\u0027s found a bug so removing it probably isn\u0027t wise.\n\nI think we should keep this until we get to a future microversion where we can remove either support for non-default quota class sets or the entire quota class set feature.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"45b5d81ccbd19a36d3e991ab34f06a8f08276f0c","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"d874b39b_540d5de5","in_reply_to":"6ba93b6c_a0707c41","updated":"2026-03-02 17:07:52.000000000","message":"so this is not a production bug correct becuase in production this woudl only be logged as a warning not a error as responce validation is only a error in ci and we warn against using it in any other context.\n\nfrom my perspectiv this isnt a cricitcal severtiy api bug as we are jsut enforcing our contract in a mode that operators have to opt into.\n\nchanging the schme to allow a value we do not suprot seams liek a regression to me.\n\nwe coudl proceed with this given the time constratis for this cycle btu i feel like this shoudl at least have a fixme comment to revert to the strict validation once we fix novaclient.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"c77489a2794e4abb3e0d379e59dcd618a0aa6acb","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"5983ce82_12f3fef2","in_reply_to":"c7767f2a_bfdb803f","updated":"2026-03-05 18:59:19.000000000","message":"Sure, if a non-default qupta class is not an error in the current API (before schema change), then we should not change that behaviour without a microversion. The only thing i wanted to check if it fail late and this schema change made it fail via response schema but that does not seems to be the case. It is 200 before and we should keep it same even non-default quota class do nothing.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"a13d3d9ca116ca3a384e38039677eb990e9aeffc","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"3dc5717c_d5150a15","in_reply_to":"d874b39b_540d5de5","updated":"2026-03-02 17:13:50.000000000","message":"by the way i agree that this is being caught in the wrong place.\nthe api shoudl return a 400 and it shoudl do that in the request validate phase not the responce phase.","commit_id":"c05c78dbffac0b101d5e06a6e5a364fec0d5e6c9"}]}