)]}'
{"releasenotes/notes/add-check-for-disable-upgrade-deployment-flag-6a7e8588be282df9.yaml":[{"author":{"_account_id":3153,"name":"Emilien Macchi","email":"emilien@redhat.com","username":"emilienm"},"change_message_id":"e41e9c04dd1e4add695f662d912fa340a355392e","unresolved":false,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"upgrade:"},{"line_number":3,"context_line":"  - This adds a check that the disable_upgrade_deployment flag is set at least"},{"line_number":4,"context_line":"    once in the deployment roles_data.yaml and otherwise LOGS a warning."}],"source_content_type":"text/x-yaml","patch_set":4,"id":"1f1a1f67_b73e7bb5","line":3,"updated":"2017-07-18 18:12:41.000000000","message":"The release note isn\u0027t a commit message, I think we should write something like:\n\nWe now check that disable_upgrade_deployment flag is set at least once in the roles_data.yaml deployment file. Otherwise, we send a warning via the logs.\n\n\nWhat do you think?","commit_id":"b844a84089f1f92742afd7560eeed086f1117f76"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"de2b78597058e7a4f11682ea502d4154815ed1b1","unresolved":false,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"upgrade:"},{"line_number":3,"context_line":"  - This adds a check that the disable_upgrade_deployment flag is set at least"},{"line_number":4,"context_line":"    once in the deployment roles_data.yaml and otherwise LOGS a warning."}],"source_content_type":"text/x-yaml","patch_set":4,"id":"1f1a1f67_f4e4a43e","line":3,"in_reply_to":"1f1a1f67_b73e7bb5","updated":"2017-07-20 14:49:33.000000000","message":"ack thanks very much - I posted an alternative review against the client in https://review.openstack.org/#/c/481635/\n\ni was waiting for more feedback about the alternatives before abandoning one of them\n\nI think I will abandon this one, and go for the client patch\n(the advantage is explained in the commit message there)\n\nIf instead folks think we should use this one then I will update as per your suggestion for the release note.\n\nthanks","commit_id":"b844a84089f1f92742afd7560eeed086f1117f76"}],"tripleo_common/actions/templates.py":[{"author":{"_account_id":16515,"name":"mbu","email":"mat.bultel@gmail.com","username":"matbu"},"change_message_id":"ac00028ebbc261ccdb92f6ef72eca0979a402163","unresolved":false,"context_lines":[{"line_number":167,"context_line":"                return True"},{"line_number":168,"context_line":"        LOG.warning(\u0027WARNING! None of the roles in the role_data have the \u0027"},{"line_number":169,"context_line":"                    \u0027disable_upgrade_deployment flag set. Please validate \u0027"},{"line_number":170,"context_line":"                    \u0027the specified role_data: %s \u0027, role_data)"},{"line_number":171,"context_line":"        if not ignore_upgrade_flag_absent:"},{"line_number":172,"context_line":"            raise RoleMetadataError(\"No disable_upgrade_deployment flag set\")"},{"line_number":173,"context_line":"        return False"}],"source_content_type":"text/x-python","patch_set":4,"id":"5f201791_7fc3f5ed","line":170,"updated":"2017-06-28 13:33:40.000000000","message":"I think, as you invite user to validate the role_data before continuing, it may be better to throw an exception here","commit_id":"b844a84089f1f92742afd7560eeed086f1117f76"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"1ad6b151494547f06d9745ccffa79cc1b8e330ec","unresolved":false,"context_lines":[{"line_number":167,"context_line":"                return True"},{"line_number":168,"context_line":"        LOG.warning(\u0027WARNING! None of the roles in the role_data have the \u0027"},{"line_number":169,"context_line":"                    \u0027disable_upgrade_deployment flag set. Please validate \u0027"},{"line_number":170,"context_line":"                    \u0027the specified role_data: %s \u0027, role_data)"},{"line_number":171,"context_line":"        if not ignore_upgrade_flag_absent:"},{"line_number":172,"context_line":"            raise RoleMetadataError(\"No disable_upgrade_deployment flag set\")"},{"line_number":173,"context_line":"        return False"}],"source_content_type":"text/x-python","patch_set":4,"id":"3f1d235d_fda01b94","line":170,"in_reply_to":"3f1d235d_26e31808","updated":"2017-07-07 13:37:17.000000000","message":"I just posted an alternative here I606a3010e7ecd83f34b852a35ed298c30f65adeb \n\nThe advantage is that it may have slightly more teeth. If the operator enables the validation warnings to be fatal with the existing flag, then we can raise the exception. wdyt?","commit_id":"b844a84089f1f92742afd7560eeed086f1117f76"},{"author":{"_account_id":8297,"name":"Sofer Athlan-Guyot","email":"sathlang@redhat.com","username":"chem"},"change_message_id":"2319587458c787a0696ab1758381e3fed430ed57","unresolved":false,"context_lines":[{"line_number":167,"context_line":"                return True"},{"line_number":168,"context_line":"        LOG.warning(\u0027WARNING! None of the roles in the role_data have the \u0027"},{"line_number":169,"context_line":"                    \u0027disable_upgrade_deployment flag set. Please validate \u0027"},{"line_number":170,"context_line":"                    \u0027the specified role_data: %s \u0027, role_data)"},{"line_number":171,"context_line":"        if not ignore_upgrade_flag_absent:"},{"line_number":172,"context_line":"            raise RoleMetadataError(\"No disable_upgrade_deployment flag set\")"},{"line_number":173,"context_line":"        return False"}],"source_content_type":"text/x-python","patch_set":4,"id":"3f1d235d_26e31808","line":170,"in_reply_to":"5f201791_7fc3f5ed","updated":"2017-06-29 13:25:23.000000000","message":"As discussed during the mtg, it may be valid to not have any disable_upgrade_deployment entry in an custom role data.  So moving this to a error would requires a command line option to be able to disable it.\n\nKeeping it as a warning may help troubleshooting for the most common case.  Anyway, it\u0027s just an heuristic, so I would rather go with a warning.","commit_id":"b844a84089f1f92742afd7560eeed086f1117f76"}]}