)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Cinder volume rollback snapshot"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"It makes sense to be able to revert a volume to a previously taken snapshot."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"blueprint: revert-volume-to-snapshot"},{"line_number":12,"context_line":"Change-Id: I88a1bd60e703bf9b434924acee7bbd74f72b1c2a"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":4,"id":"9abb7d3a_f152a4dc","line":9,"range":{"start_line":9,"start_character":67,"end_line":9,"end_character":75},"updated":"2016-05-27 17:54:58.000000000","message":"This line should get wrapped at 72 characters.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"}],"specs/newton/cinder-volume-rollback-snapshot.rst":[{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":28,"context_line":""},{"line_number":29,"context_line":"The purpose of this feature is to make user to recover instance more"},{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the asigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recovery a instance,"},{"line_number":34,"context_line":"we have the following ways:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_b670ae73","line":31,"updated":"2016-05-16 05:30:43.000000000","message":"s/asigned/assigned","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":28,"context_line":""},{"line_number":29,"context_line":"The purpose of this feature is to make user to recover instance more"},{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the asigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recovery a instance,"},{"line_number":34,"context_line":"we have the following ways:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_63d16157","line":31,"in_reply_to":"dab17558_b670ae73","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":29,"context_line":"The purpose of this feature is to make user to recover instance more"},{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the asigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recovery a instance,"},{"line_number":34,"context_line":"we have the following ways:"},{"line_number":35,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_d696124b","line":32,"updated":"2016-05-16 05:30:43.000000000","message":"may not have satisfied","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":29,"context_line":"The purpose of this feature is to make user to recover instance more"},{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the asigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recovery a instance,"},{"line_number":34,"context_line":"we have the following ways:"},{"line_number":35,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_43d6dd4f","line":32,"in_reply_to":"dab17558_d696124b","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the asigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recovery a instance,"},{"line_number":34,"context_line":"we have the following ways:"},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"The actual user workflow is:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_56898225","line":33,"updated":"2016-05-16 05:30:43.000000000","message":"s/recovery/recover","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the asigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recovery a instance,"},{"line_number":34,"context_line":"we have the following ways:"},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"The actual user workflow is:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_a3c7c996","line":33,"in_reply_to":"dab17558_56898225","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":43,"context_line":"7. User attach new volumes"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"* Use create volume from snapshot, most of vendors that this function need to"},{"line_number":47,"context_line":"  copy data to the new volume, It need to spend more time to copy the datas."},{"line_number":48,"context_line":"  But the end user hope that the instance can recover quickly."},{"line_number":49,"context_line":""},{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_3bf4a759","line":47,"range":{"start_line":46,"start_character":35,"end_line":47,"end_character":29},"updated":"2016-05-16 04:34:03.000000000","message":"I think this is in fact not at all true - most use some form of COW to avoid data copying.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"91b41ded8b0dff05f390582d1a209714732a7f4b","unresolved":false,"context_lines":[{"line_number":43,"context_line":"7. User attach new volumes"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"* Use create volume from snapshot, most of vendors that this function need to"},{"line_number":47,"context_line":"  copy data to the new volume, It need to spend more time to copy the datas."},{"line_number":48,"context_line":"  But the end user hope that the instance can recover quickly."},{"line_number":49,"context_line":""},{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"}],"source_content_type":"text/x-rst","patch_set":3,"id":"bab6814e_ff575923","line":47,"range":{"start_line":46,"start_character":35,"end_line":47,"end_character":29},"in_reply_to":"dab17558_3bf4a759","updated":"2016-05-23 08:16:36.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":49,"context_line":""},{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"},{"line_number":51,"context_line":"  UUID in the system, but several application use that to boot. If use"},{"line_number":52,"context_line":"  rollback snapshot, it will keeping the same volume id"},{"line_number":53,"context_line":"  (and therefore disk UUID) makes reverting a instance much easier since"},{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_368e7e2e","line":52,"updated":"2016-05-16 05:30:43.000000000","message":"s/keeping/keep","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":49,"context_line":""},{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"},{"line_number":51,"context_line":"  UUID in the system, but several application use that to boot. If use"},{"line_number":52,"context_line":"  rollback snapshot, it will keeping the same volume id"},{"line_number":53,"context_line":"  (and therefore disk UUID) makes reverting a instance much easier since"},{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_430b7dd6","line":52,"in_reply_to":"dab17558_368e7e2e","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"},{"line_number":51,"context_line":"  UUID in the system, but several application use that to boot. If use"},{"line_number":52,"context_line":"  rollback snapshot, it will keeping the same volume id"},{"line_number":53,"context_line":"  (and therefore disk UUID) makes reverting a instance much easier since"},{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_7684860a","line":53,"updated":"2016-05-16 05:30:43.000000000","message":"Add \"and\" before \"makes\".","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"},{"line_number":51,"context_line":"  UUID in the system, but several application use that to boot. If use"},{"line_number":52,"context_line":"  rollback snapshot, it will keeping the same volume id"},{"line_number":53,"context_line":"  (and therefore disk UUID) makes reverting a instance much easier since"},{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_a30ce9ec","line":53,"in_reply_to":"dab17558_7684860a","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":53,"context_line":"  (and therefore disk UUID) makes reverting a instance much easier since"},{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_f6707642","line":56,"updated":"2016-05-16 05:30:43.000000000","message":"Add \"to\" before \"recover\".","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":53,"context_line":"  (and therefore disk UUID) makes reverting a instance much easier since"},{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_03fe95ad","line":56,"in_reply_to":"dab17558_f6707642","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"},{"line_number":60,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_3695de32","line":57,"updated":"2016-05-16 05:30:43.000000000","message":"s/is not exist/does not exist","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":54,"context_line":"  device names inside the instance stay the same."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"},{"line_number":60,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_a3234957","line":57,"in_reply_to":"dab17558_3695de32","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"},{"line_number":60,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"},{"line_number":61,"context_line":"  find it may spend more time to copy the data to the new volume."}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_db03db40","line":58,"range":{"start_line":58,"start_character":17,"end_line":58,"end_character":24},"updated":"2016-05-16 04:34:03.000000000","message":"typo","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"},{"line_number":60,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"},{"line_number":61,"context_line":"  find it may spend more time to copy the data to the new volume."}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_a3da2949","line":58,"range":{"start_line":58,"start_character":17,"end_line":58,"end_character":24},"in_reply_to":"dab17558_56b7c2d7","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":55,"context_line":""},{"line_number":56,"context_line":"* Use restore backup, restore backup data to volume may be a way recover"},{"line_number":57,"context_line":"  data when the storage is not exist, but most time the instance can\u0027t work"},{"line_number":58,"context_line":"  may be not the sotrage\u0027s problem, the problem may only some data not"},{"line_number":59,"context_line":"  concurrency with the systems. So we also hope that the data can recover"},{"line_number":60,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"},{"line_number":61,"context_line":"  find it may spend more time to copy the data to the new volume."}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_56b7c2d7","line":58,"range":{"start_line":58,"start_character":17,"end_line":58,"end_character":24},"in_reply_to":"dab17558_db03db40","updated":"2016-05-16 05:30:43.000000000","message":"the problem may be only some data is not concurrent with the system.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":61,"context_line":"  find it may spend more time to copy the data to the new volume."},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"As mentioned above, we hope that can offer a way to recover instance quickly"},{"line_number":64,"context_line":"use an easiest way, rollback snapshot may be a good choice."},{"line_number":65,"context_line":""},{"line_number":66,"context_line":"Use Cases"},{"line_number":67,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_96e48ae0","line":64,"updated":"2016-05-16 05:30:43.000000000","message":"s/use/using","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":61,"context_line":"  find it may spend more time to copy the data to the new volume."},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"As mentioned above, we hope that can offer a way to recover instance quickly"},{"line_number":64,"context_line":"use an easiest way, rollback snapshot may be a good choice."},{"line_number":65,"context_line":""},{"line_number":66,"context_line":"Use Cases"},{"line_number":67,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_c3e5ad09","line":64,"in_reply_to":"dab17558_96e48ae0","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":67,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":68,"context_line":""},{"line_number":69,"context_line":"This is intended to be a DR function. The primary use case is for a instance"},{"line_number":70,"context_line":"falldown but the storage is also live, so roolback snapshot to the volume"},{"line_number":71,"context_line":"main user case. "},{"line_number":72,"context_line":"1. User creates a volume(s)"},{"line_number":73,"context_line":"2. User attach volume to instance"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_bb009735","line":70,"range":{"start_line":70,"start_character":42,"end_line":70,"end_character":50},"updated":"2016-05-16 04:34:03.000000000","message":"typo","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"91b41ded8b0dff05f390582d1a209714732a7f4b","unresolved":false,"context_lines":[{"line_number":67,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":68,"context_line":""},{"line_number":69,"context_line":"This is intended to be a DR function. The primary use case is for a instance"},{"line_number":70,"context_line":"falldown but the storage is also live, so roolback snapshot to the volume"},{"line_number":71,"context_line":"main user case. "},{"line_number":72,"context_line":"1. User creates a volume(s)"},{"line_number":73,"context_line":"2. User attach volume to instance"}],"source_content_type":"text/x-rst","patch_set":3,"id":"bab6814e_1f37bdff","line":70,"range":{"start_line":70,"start_character":42,"end_line":70,"end_character":50},"in_reply_to":"dab17558_bb009735","updated":"2016-05-23 08:16:36.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":74,"context_line":"3. User creates a snapshot"},{"line_number":75,"context_line":"4. Something happens causing the need of a rollback"},{"line_number":76,"context_line":"5. User dettach volume from instance"},{"line_number":77,"context_line":"6. User rollback snapshot(s)"},{"line_number":78,"context_line":""},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"Proposed change"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_9bb3139a","line":77,"range":{"start_line":77,"start_character":0,"end_line":77,"end_character":28},"updated":"2016-05-16 04:34:03.000000000","message":"You have give zero reason for why this is better than \u0027user creates a new volume from the snapshot and uses that\u0027. It\u0027s safer - the data on the faulty volume is not lost to recovery or analysis.\n\nThe idea that a specific UUID is more than a handle to a specific view of a set of data is just incorrect - any attempt to see the UUID as not changing being a win is something I\u0027m strongly against.\n\nJust because some subset of backends can\u0027t do create from snap efficiently is not a good reason to go misrepresenting the problem space - it\u0027s a good reason to avoid using that as a backend for cinder.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":86,"context_line":"::"},{"line_number":87,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"},{"line_number":88,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":89,"context_line":"    exists in cinder. If not, it would raise an exception.NotFound."},{"line_number":90,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"},{"line_number":91,"context_line":"    have related snapshot, it would raise"},{"line_number":92,"context_line":"    exception.RollbackSnapshotInvalidReference."}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_56050278","line":89,"updated":"2016-05-16 05:30:43.000000000","message":"exception.VolumeNotFound","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":86,"context_line":"::"},{"line_number":87,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"},{"line_number":88,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":89,"context_line":"    exists in cinder. If not, it would raise an exception.NotFound."},{"line_number":90,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"},{"line_number":91,"context_line":"    have related snapshot, it would raise"},{"line_number":92,"context_line":"    exception.RollbackSnapshotInvalidReference."}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_63a201b0","line":89,"in_reply_to":"dab17558_56050278","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":91,"context_line":"    have related snapshot, it would raise"},{"line_number":92,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":93,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":94,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":95,"context_line":"    "},{"line_number":96,"context_line":"    During a snapashot restoration, there are two objects that are"},{"line_number":97,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_96fbea7a","line":94,"range":{"start_line":94,"start_character":4,"end_line":94,"end_character":47},"updated":"2016-05-16 05:30:43.000000000","message":"exception.RollbackSnapshotInvalidSize is more appropriate.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":91,"context_line":"    have related snapshot, it would raise"},{"line_number":92,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":93,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":94,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":95,"context_line":"    "},{"line_number":96,"context_line":"    During a snapashot restoration, there are two objects that are"},{"line_number":97,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_a3a8698d","line":94,"range":{"start_line":94,"start_character":4,"end_line":94,"end_character":47},"in_reply_to":"dab17558_96fbea7a","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":92,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":93,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":94,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":95,"context_line":"    "},{"line_number":96,"context_line":"    During a snapashot restoration, there are two objects that are"},{"line_number":97,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":98,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_f6c23652","line":95,"updated":"2016-05-16 05:30:43.000000000","message":"white space","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":92,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":93,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":94,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":95,"context_line":"    "},{"line_number":96,"context_line":"    During a snapashot restoration, there are two objects that are"},{"line_number":97,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":98,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_83ad659d","line":95,"in_reply_to":"dab17558_f6c23652","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":94,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":95,"context_line":"    "},{"line_number":96,"context_line":"    During a snapashot restoration, there are two objects that are"},{"line_number":97,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":98,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"},{"line_number":99,"context_line":"    operations must not be allowed to either object during the"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_16c6da67","line":96,"updated":"2016-05-16 05:30:43.000000000","message":"s/snapashot/snapshot","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":94,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":95,"context_line":"    "},{"line_number":96,"context_line":"    During a snapashot restoration, there are two objects that are"},{"line_number":97,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":98,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"},{"line_number":99,"context_line":"    operations must not be allowed to either object during the"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_e396f14a","line":96,"in_reply_to":"dab17558_16c6da67","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":100,"context_line":"    restoration."},{"line_number":101,"context_line":"    "},{"line_number":102,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":103,"context_line":"    data from the rollback snapshot; After the completion of the"},{"line_number":104,"context_line":"    rollback, driver return a flag(delete_snapshot) to decide whether"},{"line_number":105,"context_line":"    all other snapshots taken after the rollback snapshot will not be"},{"line_number":106,"context_line":"    deleted."},{"line_number":107,"context_line":"    "},{"line_number":108,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":109,"context_line":"    is rollback_error. If volume is in use, it will raise an VolumeIsBusy"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_7b6f4ffc","line":106,"range":{"start_line":103,"start_character":37,"end_line":106,"end_character":12},"updated":"2016-05-16 04:34:03.000000000","message":"This gives driver specific behaviour differences to the API, which is something cinder strongly aims to avoid.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":112,"context_line":""},{"line_number":113,"context_line":"Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":114,"context_line":"driver."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"Alternatives"},{"line_number":117,"context_line":"------------"},{"line_number":118,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_1b56032a","line":115,"range":{"start_line":115,"start_character":0,"end_line":115,"end_character":0},"updated":"2016-05-16 04:34:03.000000000","message":"If this is going to be implemented, it will absolutely need some generic path implemented that works using the driver primatives we already have, while giving drivers a chance to optimise. Adding an API like this that only works for drivers that implement it is going result in painful API fragmentation.\n\nYou can probably reuse the volume migration rename dance to achieve this with relatively little effoort.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":126,"context_line":"REST API impact"},{"line_number":127,"context_line":"---------------"},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"Add a new roolback API to rollback a volume to a previous snapshot."},{"line_number":130,"context_line":"Both \"volume_id\" and \"snapshot_id\" has to be provided for"},{"line_number":131,"context_line":"the rollback API."},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_b66b8e1d","line":129,"updated":"2016-05-16 05:30:43.000000000","message":"s/roolback/rollback","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":126,"context_line":"REST API impact"},{"line_number":127,"context_line":"---------------"},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"Add a new roolback API to rollback a volume to a previous snapshot."},{"line_number":130,"context_line":"Both \"volume_id\" and \"snapshot_id\" has to be provided for"},{"line_number":131,"context_line":"the rollback API."},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_63904144","line":129,"in_reply_to":"dab17558_b66b8e1d","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":131,"context_line":"the rollback API."},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"* Rollback snapshot"},{"line_number":134,"context_line":"* /v2/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":135,"context_line":"* Method: POST"},{"line_number":136,"context_line":"* JSON schema definition for V3::"},{"line_number":137,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_36f4be59","line":134,"updated":"2016-05-16 05:30:43.000000000","message":"v3","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"b5514f844c608c1133afd5fe04a1b20b6996ad71","unresolved":false,"context_lines":[{"line_number":131,"context_line":"the rollback API."},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"* Rollback snapshot"},{"line_number":134,"context_line":"* /v2/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":135,"context_line":"* Method: POST"},{"line_number":136,"context_line":"* JSON schema definition for V3::"},{"line_number":137,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_a396a93b","line":134,"in_reply_to":"dab17558_36f4be59","updated":"2016-05-23 08:15:03.000000000","message":"Done","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":140,"context_line":"            {"},{"line_number":141,"context_line":"                \"display_name\": \"nightly001\","},{"line_number":142,"context_line":"                \"display_description\": \"test\","},{"line_number":143,"context_line":"                \"volume_id\": \"xxxxxxxx\","},{"line_number":144,"context_line":"                \"snapshot_id\": \"xxxxxxxx\","},{"line_number":145,"context_line":"                ......"},{"line_number":146,"context_line":"            }"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_76ada69a","line":143,"updated":"2016-05-16 05:30:43.000000000","message":"There\u0027s already volume_id in the URL. It is redundant to have it in the request body.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":154,"context_line":"Notifications impact"},{"line_number":155,"context_line":"--------------------"},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"None"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"Other end user impact"},{"line_number":160,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_5b324b11","line":157,"range":{"start_line":157,"start_character":0,"end_line":157,"end_character":4},"updated":"2016-05-16 04:34:03.000000000","message":"There will need to be a notification emitted for a rollback event.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":164,"context_line":"Performance Impact"},{"line_number":165,"context_line":"------------------"},{"line_number":166,"context_line":""},{"line_number":167,"context_line":"None"},{"line_number":168,"context_line":""},{"line_number":169,"context_line":"Other deployer impact"},{"line_number":170,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_db263bcd","line":167,"range":{"start_line":167,"start_character":0,"end_line":167,"end_character":4},"updated":"2016-05-16 04:34:03.000000000","message":"If this has exactly the same performance as create from snap, then the arguments for adding it are pretty weak","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"94b35612cfe723d756a8b5f553e109ea01e76925","unresolved":false,"context_lines":[{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Implementation"},{"line_number":181,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":182,"context_line":"I propose that rollback_snapshot be added as a new GET action on the"},{"line_number":183,"context_line":"os-volume-manage Cinder extension. It would be separately specifiable in"},{"line_number":184,"context_line":"Cinder\u0027s policy.json file to restrict access appropriately (default being"},{"line_number":185,"context_line":"admin_api)."},{"line_number":186,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_b6d9cef5","line":183,"range":{"start_line":182,"start_character":0,"end_line":183,"end_character":34},"updated":"2016-05-16 05:30:43.000000000","message":"An action should be a \"POST\", not a \"GET\".  Also why adding it to an existing action?  This should be a different action.  What you said here is not the same as the REST API listed above.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"5119504958723c93d4b06c5aba7e39b42b29330f","unresolved":false,"context_lines":[{"line_number":181,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":182,"context_line":"I propose that rollback_snapshot be added as a new GET action on the"},{"line_number":183,"context_line":"os-volume-manage Cinder extension. It would be separately specifiable in"},{"line_number":184,"context_line":"Cinder\u0027s policy.json file to restrict access appropriately (default being"},{"line_number":185,"context_line":"admin_api)."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Assignee(s)"},{"line_number":188,"context_line":"-----------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"dab17558_1b2d63a8","line":185,"range":{"start_line":184,"start_character":60,"end_line":185,"end_character":9},"updated":"2016-05-16 04:34:03.000000000","message":"Why this default? Admins normally have nothing to do with snapshots and similar.","commit_id":"c2deba04f9278ec12e47cc2603412bef616818df"},{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":30,"context_line":"conveniently. User can recover instance by rollback snapshot to the volume"},{"line_number":31,"context_line":"through the assigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not have satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recover a instance,"},{"line_number":34,"context_line":"we have the following ways:"},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"The actual user workflow is:"}],"source_content_type":"text/x-rst","patch_set":4,"id":"9abb7d3a_342b7a84","line":33,"range":{"start_line":33,"start_character":61,"end_line":33,"end_character":62},"updated":"2016-05-27 17:54:58.000000000","message":"a -\u003e an","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":31,"context_line":"through the assigned snapshot. But why we need to use rollback snapshot"},{"line_number":32,"context_line":"to volume function? For the current function in cinder may not have satisfied"},{"line_number":33,"context_line":"the end user\u0027s requirements. Currently if we want to recover a instance,"},{"line_number":34,"context_line":"we have the following ways:"},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"The actual user workflow is:"},{"line_number":37,"context_line":"1. User creates a volume(s)"},{"line_number":38,"context_line":"2. User attach volume to instance"},{"line_number":39,"context_line":"3. User creates a snapshot"}],"source_content_type":"text/x-rst","patch_set":4,"id":"9abb7d3a_f448d25d","line":36,"range":{"start_line":34,"start_character":5,"end_line":36,"end_character":28},"updated":"2016-05-27 17:54:58.000000000","message":"Suggest: \u0027...we have to go through the following workflow: \u0027 and then have the list.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":43,"context_line":"7. User attach new volumes"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"* Use create volume from snapshot, most of vendors that this function need to"},{"line_number":47,"context_line":"  copy data to the new volume, It need to spend more time to copy the datas."},{"line_number":48,"context_line":"  But the end user hope that the instance can recover quickly."},{"line_number":49,"context_line":""},{"line_number":50,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_56e3b205","line":47,"range":{"start_line":46,"start_character":34,"end_line":47,"end_character":29},"updated":"2016-05-24 07:09:12.000000000","message":"As before, this statement is simply not true.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":74,"context_line":"3. User creates a snapshot"},{"line_number":75,"context_line":"4. Something happens causing the need of a rollback"},{"line_number":76,"context_line":"5. User dettach volume from instance"},{"line_number":77,"context_line":"6. User rollback snapshot(s)"},{"line_number":78,"context_line":""},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"Proposed change"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_96ed1af2","line":77,"range":{"start_line":77,"start_character":3,"end_line":77,"end_character":25},"updated":"2016-05-24 07:09:12.000000000","message":"Previous comment not addressed","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":100,"context_line":"    restoration."},{"line_number":101,"context_line":""},{"line_number":102,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":103,"context_line":"    data from the rollback snapshot; After the completion of the"},{"line_number":104,"context_line":"    rollback, driver return a flag(delete_snapshot) to decide whether"},{"line_number":105,"context_line":"    all other snapshots taken after the rollback snapshot will not be"},{"line_number":106,"context_line":"    deleted."},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":109,"context_line":"    is rollback_error. If volume is in use, it will raise an VolumeIsBusy"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_36b38e0c","line":106,"range":{"start_line":103,"start_character":38,"end_line":106,"end_character":12},"updated":"2016-05-24 07:09:12.000000000","message":"Previous comment not addressed","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":112,"context_line":""},{"line_number":113,"context_line":"Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":114,"context_line":"driver."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"Alternatives"},{"line_number":117,"context_line":"------------"},{"line_number":118,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_16b80ae5","line":115,"updated":"2016-05-24 07:09:12.000000000","message":"Previous comment not addressed","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":116,"context_line":"Alternatives"},{"line_number":117,"context_line":"------------"},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"create volume from snapshot"},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Data model impact"},{"line_number":122,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"9abb7d3a_7117d451","line":119,"updated":"2016-05-27 17:54:58.000000000","message":"It might be helpful to add in here why this is not a good alternative, and why the proposed is a better approach.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":140,"context_line":"            {"},{"line_number":141,"context_line":"                \"display_name\": \"nightly001\","},{"line_number":142,"context_line":"                \"display_description\": \"test\","},{"line_number":143,"context_line":"                \"volume_id\": \"xxxxxxxx\","},{"line_number":144,"context_line":"                \"snapshot_id\": \"xxxxxxxx\","},{"line_number":145,"context_line":"                ......"},{"line_number":146,"context_line":"            }"}],"source_content_type":"text/x-rst","patch_set":4,"id":"9abb7d3a_b1ce5c58","line":143,"range":{"start_line":143,"start_character":17,"end_line":143,"end_character":40},"updated":"2016-05-27 17:54:58.000000000","message":"As Xing pointed out in a previous patchset, the vol_id is already in the URL.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":154,"context_line":"Notifications impact"},{"line_number":155,"context_line":"--------------------"},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"None"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"Other end user impact"},{"line_number":160,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"9abb7d3a_512bb0eb","line":157,"range":{"start_line":157,"start_character":0,"end_line":157,"end_character":4},"updated":"2016-05-27 17:54:58.000000000","message":"Duncan is right, there needs to be some notification about what has happened. If the rollback succeeded or not, etc.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":154,"context_line":"Notifications impact"},{"line_number":155,"context_line":"--------------------"},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"None"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"Other end user impact"},{"line_number":160,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_b6c77e63","line":157,"range":{"start_line":157,"start_character":0,"end_line":157,"end_character":4},"updated":"2016-05-24 07:09:12.000000000","message":"Previous comment not addressed","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":164,"context_line":"Performance Impact"},{"line_number":165,"context_line":"------------------"},{"line_number":166,"context_line":""},{"line_number":167,"context_line":"None"},{"line_number":168,"context_line":""},{"line_number":169,"context_line":"Other deployer impact"},{"line_number":170,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_7694f643","line":167,"range":{"start_line":167,"start_character":0,"end_line":167,"end_character":4},"updated":"2016-05-24 07:09:12.000000000","message":"Previous comment not addressed","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":16708,"name":"Kendall Nelson","display_name":"Kendall (diablo_rojo)","email":"kennelson11@gmail.com","username":"kjnelson"},"change_message_id":"9d0a3d1dd16d36768aedb5c60c4608143c13ae75","unresolved":false,"context_lines":[{"line_number":182,"context_line":"I propose that rollback_snapshot be added as a new GET action on the"},{"line_number":183,"context_line":"os-volume-manage Cinder extension. It would be separately specifiable in"},{"line_number":184,"context_line":"Cinder\u0027s policy.json file to restrict access appropriately (default being"},{"line_number":185,"context_line":"admin_api)."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Assignee(s)"},{"line_number":188,"context_line":"-----------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"9abb7d3a_5163b01a","line":185,"range":{"start_line":185,"start_character":0,"end_line":185,"end_character":9},"updated":"2016-05-27 17:54:58.000000000","message":"Again, Duncan had a good point here. Admin doesn\u0027t make sense given that it isn\u0027t normally Admin\u0027s making snapshots of volumes, its the users.","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"a8762de4f202b4a36e80574b353732f4d6c4ff79","unresolved":false,"context_lines":[{"line_number":182,"context_line":"I propose that rollback_snapshot be added as a new GET action on the"},{"line_number":183,"context_line":"os-volume-manage Cinder extension. It would be separately specifiable in"},{"line_number":184,"context_line":"Cinder\u0027s policy.json file to restrict access appropriately (default being"},{"line_number":185,"context_line":"admin_api)."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Assignee(s)"},{"line_number":188,"context_line":"-----------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bab6814e_f6862615","line":185,"range":{"start_line":185,"start_character":0,"end_line":185,"end_character":9},"updated":"2016-05-24 07:09:12.000000000","message":"Previous comment not addressed","commit_id":"887b2e56b008436ae84ccecf4dc27a7cdde4dab6"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"ba66474aa753f5adfdc43b8ac21497a69eae2809","unresolved":false,"context_lines":[{"line_number":78,"context_line":""},{"line_number":79,"context_line":"We already have create a new share from snapshot function, Why do we need"},{"line_number":80,"context_line":"rollback snapshot?"},{"line_number":81,"context_line":"* Rollback snapshot not consumes the user quotas. Solve the problem that"},{"line_number":82,"context_line":"  volume can not be recovered when the quota filled."},{"line_number":83,"context_line":"* Some of vendors use rollback snapshot funciton don\u0027t need to copy data"},{"line_number":84,"context_line":"  to the new volume. It will save much time."},{"line_number":85,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_6fe4120d","line":82,"range":{"start_line":81,"start_character":2,"end_line":82,"end_character":51},"updated":"2016-05-27 16:33:30.000000000","message":"If you\u0027re that tight for quota, delete the \u0027problem\u0027 volume and recreate it from snapshot - zero extra quota used.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"aeec87c6cc06586d1a6b6818110e3c1609d58898","unresolved":false,"context_lines":[{"line_number":78,"context_line":""},{"line_number":79,"context_line":"We already have create a new share from snapshot function, Why do we need"},{"line_number":80,"context_line":"rollback snapshot?"},{"line_number":81,"context_line":"* Rollback snapshot not consumes the user quotas. Solve the problem that"},{"line_number":82,"context_line":"  volume can not be recovered when the quota filled."},{"line_number":83,"context_line":"* Some of vendors use rollback snapshot funciton don\u0027t need to copy data"},{"line_number":84,"context_line":"  to the new volume. It will save much time."},{"line_number":85,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_e15eff10","line":82,"range":{"start_line":81,"start_character":2,"end_line":82,"end_character":51},"in_reply_to":"9abb7d3a_640a4b23","updated":"2016-05-31 07:02:30.000000000","message":"Yup, ok, you\u0027re right.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"5b9727aa5ad72ed5952b33a02d73b1322a63d43d","unresolved":false,"context_lines":[{"line_number":78,"context_line":""},{"line_number":79,"context_line":"We already have create a new share from snapshot function, Why do we need"},{"line_number":80,"context_line":"rollback snapshot?"},{"line_number":81,"context_line":"* Rollback snapshot not consumes the user quotas. Solve the problem that"},{"line_number":82,"context_line":"  volume can not be recovered when the quota filled."},{"line_number":83,"context_line":"* Some of vendors use rollback snapshot funciton don\u0027t need to copy data"},{"line_number":84,"context_line":"  to the new volume. It will save much time."},{"line_number":85,"context_line":"* Use create volume from snapshot, the new volume could not keep the same"}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_640a4b23","line":82,"range":{"start_line":81,"start_character":2,"end_line":82,"end_character":51},"in_reply_to":"9abb7d3a_6fe4120d","updated":"2016-05-30 10:09:36.000000000","message":"The \u0027problem\u0027 volume will not be deleted when the snapshot will be used to create a new volume. \nIn other words, if we delete the \u0027problem\u0027 volume, the volume corresponding to the snapshots will be deleted.\n\nso, delete the \u0027problem\u0027 volume and recreate it from snapshot, it could not be come true in some driver.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"ba66474aa753f5adfdc43b8ac21497a69eae2809","unresolved":false,"context_lines":[{"line_number":96,"context_line":"Add rollback snapshot API."},{"line_number":97,"context_line":""},{"line_number":98,"context_line":"::"},{"line_number":99,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e [--lossy]"},{"line_number":100,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":101,"context_line":"    exists in cinder. If not, it would raise an exception.VolumeNotFound."},{"line_number":102,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_8fce9681","line":99,"range":{"start_line":99,"start_character":43,"end_line":99,"end_character":48},"updated":"2016-05-27 16:33:30.000000000","message":"This is a better suggestion than letting the driver decide for sure.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"ba66474aa753f5adfdc43b8ac21497a69eae2809","unresolved":false,"context_lines":[{"line_number":125,"context_line":"Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":126,"context_line":"driver."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Vendors can implement the optimized rollback in their drivers in order to"},{"line_number":129,"context_line":"rollback snapshot to volume without requiring create a new volume from snapshot."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"Add driver interfaces::"},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_4fe78ee4","line":129,"range":{"start_line":128,"start_character":0,"end_line":129,"end_character":79},"updated":"2016-05-27 16:33:30.000000000","message":"Sorry, not quite what I meant. If we add the tenant facing API \u0027revert volume from snapshot\u0027 then the feature needs to be implemented in such a way that it works immediately for every driver (i.e. using the existing driver entry calls and renaming the volume internally). Drivers that can optimise are then free to do so. It is *not* ok to add this feature in such a way it only ends up working on some backends and not others.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"83f46b771478ee04c031b2956aecf4f57989b1b1","unresolved":false,"context_lines":[{"line_number":125,"context_line":"Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":126,"context_line":"driver."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Vendors can implement the optimized rollback in their drivers in order to"},{"line_number":129,"context_line":"rollback snapshot to volume without requiring create a new volume from snapshot."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"Add driver interfaces::"},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"7aa08908_caaccfa2","line":129,"range":{"start_line":128,"start_character":0,"end_line":129,"end_character":79},"in_reply_to":"9abb7d3a_21864765","updated":"2016-06-07 07:44:26.000000000","message":"If the rollback snapshot feature like the migration code, it have a generic implementation. The generic implementation will be full data copy from snapshot to source volume(1. attach snapshot to host, 2. data copy from snapshot to source volume, 3. dettach snapshot).  Whether this method is reasonable? It sounds like repeat operation.\n\n\nHow about the rollback snapshot feature just like create volume from snapshot feature code, If driver does not support it, it will return false, and will remind user does not support, but volume status is still available.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"5b9727aa5ad72ed5952b33a02d73b1322a63d43d","unresolved":false,"context_lines":[{"line_number":125,"context_line":"Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":126,"context_line":"driver."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Vendors can implement the optimized rollback in their drivers in order to"},{"line_number":129,"context_line":"rollback snapshot to volume without requiring create a new volume from snapshot."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"Add driver interfaces::"},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_99ca3119","line":129,"range":{"start_line":128,"start_character":0,"end_line":129,"end_character":79},"in_reply_to":"9abb7d3a_4fe78ee4","updated":"2016-05-30 10:09:36.000000000","message":"Hi Duncan Thomas, the rollback snapshot feature just like create volume from snapshot feature, If driver does not support it, it will return false, and will remind user does not support, but volume status is still available. \nIf the rollback snapshot feature like migrate feature, it have a generic implemented. it will be data copy from snapshot to source volume. Is this will be better? Does I I understand it correctly?  Thanks Duncan Thomas","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"aeec87c6cc06586d1a6b6818110e3c1609d58898","unresolved":false,"context_lines":[{"line_number":125,"context_line":"Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":126,"context_line":"driver."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Vendors can implement the optimized rollback in their drivers in order to"},{"line_number":129,"context_line":"rollback snapshot to volume without requiring create a new volume from snapshot."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"Add driver interfaces::"},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_21864765","line":129,"range":{"start_line":128,"start_character":0,"end_line":129,"end_character":79},"in_reply_to":"9abb7d3a_99ca3119","updated":"2016-05-31 07:02:30.000000000","message":"Yes, there should be a generic implementation like the migration code.","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"aeec87c6cc06586d1a6b6818110e3c1609d58898","unresolved":false,"context_lines":[{"line_number":180,"context_line":"Notifications impact"},{"line_number":181,"context_line":"--------------------"},{"line_number":182,"context_line":""},{"line_number":183,"context_line":"None"},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"Other end user impact"},{"line_number":186,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":5,"id":"9abb7d3a_c1977b8e","line":183,"range":{"start_line":183,"start_character":0,"end_line":183,"end_character":4},"updated":"2016-05-31 07:02:30.000000000","message":"A notification should be generated on rollback","commit_id":"320f4c0bb46b64758c7f58d3586af9dea274b0f9"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"af384b26e4cd628b2bfe7f9aa76490a8d78c4762","unresolved":false,"context_lines":[{"line_number":142,"context_line":"        :param context: Our working context."},{"line_number":143,"context_line":"        :param volume: the volume to be rollbacked."},{"line_number":144,"context_line":"        :param snapshot: the snapshot data rollback to volume."},{"line_number":145,"context_line":"        :param lossy: whether all other snapshots taken after the rollback"},{"line_number":146,"context_line":"        snapshot will not be deleted, the default is False."},{"line_number":147,"context_line":"        raise NotImplementedError()"},{"line_number":148,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"7aa08908_b28bff67","line":145,"range":{"start_line":145,"start_character":0,"end_line":145,"end_character":74},"updated":"2016-06-15 15:51:54.000000000","message":"Manila decided to only allow revert to the latest snapshot (so that you have to delete the latest first to get to an older snapshot) instead of allowing \"lossy\" reverts where deletes are done implicitly.  I\u0027m concerned that different projects are taking quite different approaches to similar problems without any explicit argument for doing things differently.","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"2feb33a7a03e850d6293baf04d3692851ee0ccc5","unresolved":false,"context_lines":[{"line_number":142,"context_line":"        :param context: Our working context."},{"line_number":143,"context_line":"        :param volume: the volume to be rollbacked."},{"line_number":144,"context_line":"        :param snapshot: the snapshot data rollback to volume."},{"line_number":145,"context_line":"        :param lossy: whether all other snapshots taken after the rollback"},{"line_number":146,"context_line":"        snapshot will not be deleted, the default is False."},{"line_number":147,"context_line":"        raise NotImplementedError()"},{"line_number":148,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"5a9d85d2_6a77eed4","line":145,"range":{"start_line":145,"start_character":0,"end_line":145,"end_character":74},"in_reply_to":"7aa08908_b28bff67","updated":"2016-06-20 02:27:23.000000000","message":"Good catch, we could discuss it on IRC meeting. And add a description.","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"af384b26e4cd628b2bfe7f9aa76490a8d78c4762","unresolved":false,"context_lines":[{"line_number":150,"context_line":"------------"},{"line_number":151,"context_line":""},{"line_number":152,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":153,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Data model impact"},{"line_number":156,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":7,"id":"7aa08908_32b70ff7","line":153,"range":{"start_line":153,"start_character":32,"end_line":153,"end_character":74},"updated":"2016-06-15 15:51:54.000000000","message":"Speaking of quotas, if the volume in hand was extended since the snapshot, do you shrink it when reverting?  And if so, do you adjust quota usage down correspondingly?","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"2feb33a7a03e850d6293baf04d3692851ee0ccc5","unresolved":false,"context_lines":[{"line_number":150,"context_line":"------------"},{"line_number":151,"context_line":""},{"line_number":152,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":153,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Data model impact"},{"line_number":156,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":7,"id":"5a9d85d2_4aadd230","line":153,"range":{"start_line":153,"start_character":32,"end_line":153,"end_character":74},"in_reply_to":"7aa08908_32b70ff7","updated":"2016-06-20 02:27:23.000000000","message":"If the volume in hand was extended since the snapshot, I will shrink it when reverting. Yes, I will adjust quota usage.","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"af384b26e4cd628b2bfe7f9aa76490a8d78c4762","unresolved":false,"context_lines":[{"line_number":167,"context_line":"* Rollback snapshot"},{"line_number":168,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":169,"context_line":"* Method: POST"},{"line_number":170,"context_line":"* JSON schema definition for V3::"},{"line_number":171,"context_line":""},{"line_number":172,"context_line":"        {"},{"line_number":173,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":7,"id":"7aa08908_777229e5","line":170,"updated":"2016-06-15 15:51:54.000000000","message":"Since you work in manila as well and are likely familiar with https://review.openstack.org/#/c/315695/ I am curious whether there is a reason to prefer this api to the \u0027revert\u0027 api presented there (with volumes instead of shares of course)?","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"2feb33a7a03e850d6293baf04d3692851ee0ccc5","unresolved":false,"context_lines":[{"line_number":167,"context_line":"* Rollback snapshot"},{"line_number":168,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":169,"context_line":"* Method: POST"},{"line_number":170,"context_line":"* JSON schema definition for V3::"},{"line_number":171,"context_line":""},{"line_number":172,"context_line":"        {"},{"line_number":173,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":7,"id":"5a9d85d2_855fad0d","line":170,"in_reply_to":"7aa08908_777229e5","updated":"2016-06-20 02:27:23.000000000","message":"I am not sure I understand is correct. Does this means why not use manila \u0027revert\u0027 api method instead of this api?","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"af384b26e4cd628b2bfe7f9aa76490a8d78c4762","unresolved":false,"context_lines":[{"line_number":198,"context_line":"------------------"},{"line_number":199,"context_line":""},{"line_number":200,"context_line":"The existing create-volume-from-snapshot workflow must be protected with"},{"line_number":201,"context_line":"locks to ensure no races occur."},{"line_number":202,"context_line":""},{"line_number":203,"context_line":""},{"line_number":204,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":7,"id":"7aa08908_9793555c","line":201,"updated":"2016-06-15 15:51:54.000000000","message":"In which services?  c-vol?  c-api?  Will these need to be distributed locks?","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"2feb33a7a03e850d6293baf04d3692851ee0ccc5","unresolved":false,"context_lines":[{"line_number":198,"context_line":"------------------"},{"line_number":199,"context_line":""},{"line_number":200,"context_line":"The existing create-volume-from-snapshot workflow must be protected with"},{"line_number":201,"context_line":"locks to ensure no races occur."},{"line_number":202,"context_line":""},{"line_number":203,"context_line":""},{"line_number":204,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":7,"id":"5a9d85d2_4a231206","line":201,"in_reply_to":"7aa08908_9793555c","updated":"2016-06-20 02:27:23.000000000","message":"Done","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"af384b26e4cd628b2bfe7f9aa76490a8d78c4762","unresolved":false,"context_lines":[{"line_number":230,"context_line":"- Create driver interface"},{"line_number":231,"context_line":"- Implement LVM reference"},{"line_number":232,"context_line":"- Implement tempest test"},{"line_number":233,"context_line":""},{"line_number":234,"context_line":"Dependencies"},{"line_number":235,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":236,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"7aa08908_172925f3","line":233,"updated":"2016-06-15 15:51:54.000000000","message":"Is there cinder-client or UI work that needs to be done?\n\nDocumentation?","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"2feb33a7a03e850d6293baf04d3692851ee0ccc5","unresolved":false,"context_lines":[{"line_number":230,"context_line":"- Create driver interface"},{"line_number":231,"context_line":"- Implement LVM reference"},{"line_number":232,"context_line":"- Implement tempest test"},{"line_number":233,"context_line":""},{"line_number":234,"context_line":"Dependencies"},{"line_number":235,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":236,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"5a9d85d2_2aab268b","line":233,"in_reply_to":"7aa08908_172925f3","updated":"2016-06-20 02:27:23.000000000","message":"Done,\n\n\nDocumentation: In line 245(Documentation Impact)","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"af384b26e4cd628b2bfe7f9aa76490a8d78c4762","unresolved":false,"context_lines":[{"line_number":239,"context_line":"Testing"},{"line_number":240,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":241,"context_line":""},{"line_number":242,"context_line":"Standard unit tests and manual testing, as well as tempest test for these"},{"line_number":243,"context_line":"proposed APIs."},{"line_number":244,"context_line":""},{"line_number":245,"context_line":"Documentation Impact"}],"source_content_type":"text/x-rst","patch_set":7,"id":"7aa08908_f71459a0","line":242,"range":{"start_line":242,"start_character":0,"end_line":242,"end_character":8},"updated":"2016-06-15 15:51:54.000000000","message":"Are the non-standard unit tests :)","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"2feb33a7a03e850d6293baf04d3692851ee0ccc5","unresolved":false,"context_lines":[{"line_number":239,"context_line":"Testing"},{"line_number":240,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":241,"context_line":""},{"line_number":242,"context_line":"Standard unit tests and manual testing, as well as tempest test for these"},{"line_number":243,"context_line":"proposed APIs."},{"line_number":244,"context_line":""},{"line_number":245,"context_line":"Documentation Impact"}],"source_content_type":"text/x-rst","patch_set":7,"id":"5a9d85d2_eaa0fea7","line":242,"range":{"start_line":242,"start_character":0,"end_line":242,"end_character":8},"in_reply_to":"7aa08908_f71459a0","updated":"2016-06-20 02:27:23.000000000","message":"Done","commit_id":"2e8dbd461ee8369a42b6c3f07632ee8270e7a4ed"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"46978e668ae41c1bbb332bd6691fa6d639f19853","unresolved":false,"context_lines":[{"line_number":148,"context_line":"        implement rollback snapshot, you may need to lossy some snapshot that"},{"line_number":149,"context_line":"        taken after the rollback snapshot. Otherwise, the rollback snapshot"},{"line_number":150,"context_line":"        feature will fail, the original volume status is still available.)"},{"line_number":151,"context_line":"        raise NotImplementedError()"},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Alternatives"},{"line_number":154,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_08af0c95","line":151,"updated":"2016-07-19 12:40:01.000000000","message":"What about the alternative of allowing only rollback to the lastest snapshot instead of allowing lossy snapshots?","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":11865,"name":"Clinton Knight","email":"cknight@netapp.com","username":"cknight"},"change_message_id":"c72165a4ec9fed485e980b5a158d3ca29c6687f4","unresolved":false,"context_lines":[{"line_number":148,"context_line":"        implement rollback snapshot, you may need to lossy some snapshot that"},{"line_number":149,"context_line":"        taken after the rollback snapshot. Otherwise, the rollback snapshot"},{"line_number":150,"context_line":"        feature will fail, the original volume status is still available.)"},{"line_number":151,"context_line":"        raise NotImplementedError()"},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Alternatives"},{"line_number":154,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_760942dd","line":151,"in_reply_to":"dada55a8_08af0c95","updated":"2016-07-19 15:23:22.000000000","message":"+1  Agree with Tom \u0026 Duncan.  Deleting newer snapshots is a form of data loss.","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"46978e668ae41c1bbb332bd6691fa6d639f19853","unresolved":false,"context_lines":[{"line_number":171,"context_line":"* Rollback snapshot"},{"line_number":172,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":173,"context_line":"* Method: POST"},{"line_number":174,"context_line":"* JSON schema definition for V3::"},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"        {"},{"line_number":177,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_684048a6","line":174,"updated":"2016-07-19 12:40:01.000000000","message":"I think we produce a poor user experience doing \"rollback\", and allowing lossy rollbacks, in cinder but doing \"revert\", only to the last snapshot, without loss, in manila.","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":11865,"name":"Clinton Knight","email":"cknight@netapp.com","username":"cknight"},"change_message_id":"c72165a4ec9fed485e980b5a158d3ca29c6687f4","unresolved":false,"context_lines":[{"line_number":171,"context_line":"* Rollback snapshot"},{"line_number":172,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":173,"context_line":"* Method: POST"},{"line_number":174,"context_line":"* JSON schema definition for V3::"},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"        {"},{"line_number":177,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_768ca235","line":174,"in_reply_to":"dada55a8_684048a6","updated":"2016-07-19 15:23:22.000000000","message":"+1  Why reinvent the wheel?  You\u0027re welcome to copy the spec \u0026 code from manila.\n\nhttps://github.com/openstack/manila-specs/blob/master/specs/newton/manila-share-revert-to-snapshot.rst\nhttps://review.openstack.org/#/c/340502/\nhttps://review.openstack.org/#/c/340556/","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"5a89d6e0acd1e2f19ce7ae93ed54b833d8dc7db4","unresolved":false,"context_lines":[{"line_number":171,"context_line":"* Rollback snapshot"},{"line_number":172,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":173,"context_line":"* Method: POST"},{"line_number":174,"context_line":"* JSON schema definition for V3::"},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"        {"},{"line_number":177,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_f52431c7","line":174,"in_reply_to":"dada55a8_684048a6","updated":"2016-07-23 16:01:11.000000000","message":"Done","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"47cf93c02c498f33206e9ccee8b203163e18685e","unresolved":false,"context_lines":[{"line_number":171,"context_line":"* Rollback snapshot"},{"line_number":172,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":173,"context_line":"* Method: POST"},{"line_number":174,"context_line":"* JSON schema definition for V3::"},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"        {"},{"line_number":177,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_2bb9462f","line":174,"in_reply_to":"dada55a8_7517a154","updated":"2016-07-23 17:35:26.000000000","message":"The aim of cinder is *not* to expose every combination of vendor features. That way lies madness and unusability. We pick a sane set of well supported primitives, and carefully define behaviour around them. We\u0027re far better off not supporting some feature than adding complex, poorly defined or difficult-to-universally-support semantics.","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":15100,"name":"zhongjun","email":"jun.zhongjun2@gmail.com","username":"zhongjun2"},"change_message_id":"5a89d6e0acd1e2f19ce7ae93ed54b833d8dc7db4","unresolved":false,"context_lines":[{"line_number":171,"context_line":"* Rollback snapshot"},{"line_number":172,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":173,"context_line":"* Method: POST"},{"line_number":174,"context_line":"* JSON schema definition for V3::"},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"        {"},{"line_number":177,"context_line":"            \"rollback\":"}],"source_content_type":"text/x-rst","patch_set":8,"id":"dada55a8_7517a154","line":174,"in_reply_to":"dada55a8_768ca235","updated":"2016-07-23 16:01:11.000000000","message":"haha, you are a nice guy, thanks. \nBecause some of the vendor driver could not loss the snapshot data when user use the rollback snapshot feature. So this spec try to find a suitable way to fit for these different implement in driver.","commit_id":"30c319bc8648f5dc7eb1bb804a63eea64d3fdf14"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2ef5a49038f98c87cf1c1bde6603b12f8d8277f8","unresolved":false,"context_lines":[{"line_number":35,"context_line":"we have the following ways:"},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"The actual user workflow is:"},{"line_number":38,"context_line":"1. User creates a volume(s)"},{"line_number":39,"context_line":"2. User attach volume to instance"},{"line_number":40,"context_line":"3. User creates a snapshot"},{"line_number":41,"context_line":"4. Something happens causing the need of a rollback"}],"source_content_type":"text/x-rst","patch_set":10,"id":"1ac06dbe_b8839c03","line":38,"range":{"start_line":38,"start_character":0,"end_line":38,"end_character":27},"updated":"2016-08-22 12:30:57.000000000","message":"there should be a blank line before to make the list works if this is your intention","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2ef5a49038f98c87cf1c1bde6603b12f8d8277f8","unresolved":false,"context_lines":[{"line_number":105,"context_line":""},{"line_number":106,"context_line":"By supporting rollback snapshot, we need to do the following changes."},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"Add rollback snapshot API."},{"line_number":109,"context_line":""},{"line_number":110,"context_line":"::"},{"line_number":111,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"}],"source_content_type":"text/x-rst","patch_set":10,"id":"1ac06dbe_f826b4cc","line":108,"range":{"start_line":108,"start_character":0,"end_line":108,"end_character":26},"updated":"2016-08-22 12:30:57.000000000","message":"I think the items below:\n- Add rollbakc snapshot API...\n- Add rollback_snapshot...\n- Add driver interfaces...\n- .....\n\nshould be the docorated with same symbol,eg:*/-/+","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"444437ab749d5f40d8c35e404b62702f18dda265","unresolved":false,"context_lines":[{"line_number":131,"context_line":"    data from the rollback snapshot;"},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":134,"context_line":"    is rollback_error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":135,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":136,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":137,"context_line":""}],"source_content_type":"text/x-rst","patch_set":10,"id":"1ac06dbe_081fe3b7","line":134,"updated":"2016-08-22 20:24:22.000000000","message":"We should not introduce a new \"rollback_error\" status.  This should just be \"error\" and we should fire an async error message indicating what the error is.","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"92016b6120cf4eeecef7944906b3100e399af57c","unresolved":false,"context_lines":[{"line_number":131,"context_line":"    data from the rollback snapshot;"},{"line_number":132,"context_line":""},{"line_number":133,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":134,"context_line":"    is rollback_error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":135,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":136,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":137,"context_line":""}],"source_content_type":"text/x-rst","patch_set":10,"id":"fa7ab95a_7e46e37b","line":134,"in_reply_to":"1ac06dbe_081fe3b7","updated":"2016-08-30 11:47:46.000000000","message":"done~\nthe status changed to error","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2ef5a49038f98c87cf1c1bde6603b12f8d8277f8","unresolved":false,"context_lines":[{"line_number":155,"context_line":"the framwork will use the general way to implement the rollback snapshot"},{"line_number":156,"context_line":"feature."},{"line_number":157,"context_line":""},{"line_number":158,"context_line":"1.create a snapshot from volume"},{"line_number":159,"context_line":"2.mount snapshot to host"},{"line_number":160,"context_line":"3.copy date from snapshot to original volume"},{"line_number":161,"context_line":"4.delete the original volume"}],"source_content_type":"text/x-rst","patch_set":10,"id":"1ac06dbe_38ddec0e","line":158,"range":{"start_line":158,"start_character":0,"end_line":158,"end_character":31},"updated":"2016-08-22 12:30:57.000000000","message":"please add a blank space after the symbol of dot for each item","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2ef5a49038f98c87cf1c1bde6603b12f8d8277f8","unresolved":false,"context_lines":[{"line_number":163,"context_line":"If the cinder driver can not support mount snapshot, the general way"},{"line_number":164,"context_line":"will be the following steps."},{"line_number":165,"context_line":""},{"line_number":166,"context_line":"1.create a hidden volume"},{"line_number":167,"context_line":"2.mount hidden volume to host"},{"line_number":168,"context_line":"3.copy date from hidden volume to original volume"},{"line_number":169,"context_line":"4.delete the original volume"}],"source_content_type":"text/x-rst","patch_set":10,"id":"1ac06dbe_7805a478","line":166,"range":{"start_line":166,"start_character":0,"end_line":166,"end_character":24},"updated":"2016-08-22 12:30:57.000000000","message":"please add a blank space after the symbol of dot  for each item","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"444437ab749d5f40d8c35e404b62702f18dda265","unresolved":false,"context_lines":[{"line_number":188,"context_line":"the rollback API."},{"line_number":189,"context_line":""},{"line_number":190,"context_line":"* Rollback snapshot"},{"line_number":191,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":192,"context_line":"* Method: POST"},{"line_number":193,"context_line":"* JSON schema definition for V3::"},{"line_number":194,"context_line":""}],"source_content_type":"text/x-rst","patch_set":10,"id":"1ac06dbe_c8a5cba2","line":191,"updated":"2016-08-22 20:24:22.000000000","message":"Contains \\u200b zero-width space characters... please clean up.","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"92016b6120cf4eeecef7944906b3100e399af57c","unresolved":false,"context_lines":[{"line_number":188,"context_line":"the rollback API."},{"line_number":189,"context_line":""},{"line_number":190,"context_line":"* Rollback snapshot"},{"line_number":191,"context_line":"* /v3/​{tenant_id}​/volumes/​{volume_id}​/rollback"},{"line_number":192,"context_line":"* Method: POST"},{"line_number":193,"context_line":"* JSON schema definition for V3::"},{"line_number":194,"context_line":""}],"source_content_type":"text/x-rst","patch_set":10,"id":"fa7ab95a_3ec29be9","line":191,"in_reply_to":"1ac06dbe_c8a5cba2","updated":"2016-08-30 11:47:46.000000000","message":"done~","commit_id":"c3e55b42bf5ee0ef5af5675aee7e3293323e0e26"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":109,"context_line":"Add rollback snapshot API."},{"line_number":110,"context_line":""},{"line_number":111,"context_line":"::"},{"line_number":112,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"},{"line_number":113,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":114,"context_line":"    exists in cinder. If not, it would raise an exception.VolumeNotFound."},{"line_number":115,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_97b12192","line":112,"range":{"start_line":112,"start_character":4,"end_line":112,"end_character":39},"updated":"2016-10-06 19:09:38.000000000","message":"I think that we can use the same syntax for snapshot commands to keep consistency:\n\ncinder snapshot-rollback \u003cvolume\u003e\n\nAlso, you don\u0027t need the snapshot information as you are only reverting to the last snapshot so, if the volume have \u003e\u003d1 snapshots, it will revert to the last, otherwise it will return an error.","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":109,"context_line":"Add rollback snapshot API."},{"line_number":110,"context_line":""},{"line_number":111,"context_line":"::"},{"line_number":112,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"},{"line_number":113,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":114,"context_line":"    exists in cinder. If not, it would raise an exception.VolumeNotFound."},{"line_number":115,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_cf37484b","line":112,"range":{"start_line":112,"start_character":4,"end_line":112,"end_character":39},"in_reply_to":"1a95cdbc_97b12192","updated":"2016-10-09 02:54:43.000000000","message":"wow~,that\u0027s straight to the point, and it\u0027s will be much simplier and clearer，thanks","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":110,"context_line":""},{"line_number":111,"context_line":"::"},{"line_number":112,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"},{"line_number":113,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":114,"context_line":"    exists in cinder. If not, it would raise an exception.VolumeNotFound."},{"line_number":115,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"},{"line_number":116,"context_line":"    have related snapshot, it would raise"},{"line_number":117,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":118,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":119,"context_line":"    exception.RollbackSnapshotInvalidSize."},{"line_number":120,"context_line":"    If the snapshot is not the recent volume snapshot known to cinder, it"},{"line_number":121,"context_line":"    would raise exception.RollbackSnapshotInvalidReference."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":"    During a snapshot restoration, there are two objects that are"},{"line_number":124,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_573cf9e1","line":121,"range":{"start_line":113,"start_character":4,"end_line":121,"end_character":59},"updated":"2016-10-06 19:09:38.000000000","message":"This wouldn\u0027t be needed if you use the syntax I mentioned.","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":110,"context_line":""},{"line_number":111,"context_line":"::"},{"line_number":112,"context_line":"    cinder rollback \u003cvolume\u003e \u003csnapshot\u003e"},{"line_number":113,"context_line":"    Rollback snapshot would check whether the snapshot\u0027s parent volume"},{"line_number":114,"context_line":"    exists in cinder. If not, it would raise an exception.VolumeNotFound."},{"line_number":115,"context_line":"    If the snapshot doesn\u0027t exist in volume backend or the volume doesn\u0027t"},{"line_number":116,"context_line":"    have related snapshot, it would raise"},{"line_number":117,"context_line":"    exception.RollbackSnapshotInvalidReference."},{"line_number":118,"context_line":"    If the size of a volume not match with snapshot size, it would raise"},{"line_number":119,"context_line":"    exception.RollbackSnapshotInvalidSize."},{"line_number":120,"context_line":"    If the snapshot is not the recent volume snapshot known to cinder, it"},{"line_number":121,"context_line":"    would raise exception.RollbackSnapshotInvalidReference."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":"    During a snapshot restoration, there are two objects that are"},{"line_number":124,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_ef4244ad","line":121,"range":{"start_line":113,"start_character":4,"end_line":121,"end_character":59},"in_reply_to":"1a95cdbc_573cf9e1","updated":"2016-10-09 02:54:43.000000000","message":"Done","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":122,"context_line":""},{"line_number":123,"context_line":"    During a snapshot restoration, there are two objects that are"},{"line_number":124,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":125,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"},{"line_number":126,"context_line":"    operations must not be allowed to either object during the"},{"line_number":127,"context_line":"    restoration."},{"line_number":128,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_4b1e1560","line":125,"range":{"start_line":125,"start_character":4,"end_line":125,"end_character":15},"updated":"2016-10-06 19:09:38.000000000","message":"rollingback","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":122,"context_line":""},{"line_number":123,"context_line":"    During a snapshot restoration, there are two objects that are"},{"line_number":124,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":125,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"},{"line_number":126,"context_line":"    operations must not be allowed to either object during the"},{"line_number":127,"context_line":"    restoration."},{"line_number":128,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_eb18c177","line":125,"range":{"start_line":125,"start_character":47,"end_line":125,"end_character":58},"updated":"2016-10-06 19:09:38.000000000","message":"rollingback","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":122,"context_line":""},{"line_number":123,"context_line":"    During a snapshot restoration, there are two objects that are"},{"line_number":124,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":125,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"},{"line_number":126,"context_line":"    operations must not be allowed to either object during the"},{"line_number":127,"context_line":"    restoration."},{"line_number":128,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_8fd9c059","line":125,"range":{"start_line":125,"start_character":4,"end_line":125,"end_character":15},"in_reply_to":"1a95cdbc_4b1e1560","updated":"2016-10-09 02:54:43.000000000","message":"Done","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":122,"context_line":""},{"line_number":123,"context_line":"    During a snapshot restoration, there are two objects that are"},{"line_number":124,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":125,"context_line":"    rollbacking, the snapshot\u0027s status is also rollbacking. Other"},{"line_number":126,"context_line":"    operations must not be allowed to either object during the"},{"line_number":127,"context_line":"    restoration."},{"line_number":128,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_afdcbc46","line":125,"range":{"start_line":125,"start_character":47,"end_line":125,"end_character":58},"in_reply_to":"1a95cdbc_eb18c177","updated":"2016-10-09 02:54:43.000000000","message":"Done","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":127,"context_line":"    restoration."},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"    If the cinder driver can not support the rollback feature, it will"},{"line_number":130,"context_line":"    use the general way to run rollback feature."},{"line_number":131,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":132,"context_line":"    data from the rollback snapshot;"},{"line_number":133,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_4b7a9578","line":130,"range":{"start_line":130,"start_character":12,"end_line":130,"end_character":19},"updated":"2016-10-06 19:09:38.000000000","message":"generic implementation","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":127,"context_line":"    restoration."},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"    If the cinder driver can not support the rollback feature, it will"},{"line_number":130,"context_line":"    use the general way to run rollback feature."},{"line_number":131,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":132,"context_line":"    data from the rollback snapshot;"},{"line_number":133,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_2ff18ccd","line":130,"range":{"start_line":130,"start_character":12,"end_line":130,"end_character":19},"in_reply_to":"1a95cdbc_4b7a9578","updated":"2016-10-09 02:54:43.000000000","message":"Done","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":132,"context_line":"    data from the rollback snapshot;"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":135,"context_line":"    is error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":136,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":137,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":138,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_6bebd178","line":135,"range":{"start_line":135,"start_character":14,"end_line":135,"end_character":33},"updated":"2016-10-06 19:09:38.000000000","message":"This is confusing. It gives the impression that it will be allowed to rollback in-use snapshots. Other point, volume-api (not manager)is the point where this should be blocked.","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":131,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":132,"context_line":"    data from the rollback snapshot;"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":135,"context_line":"    is error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":136,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":137,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":138,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_8bfbbdc0","line":135,"range":{"start_line":134,"start_character":4,"end_line":135,"end_character":14},"updated":"2016-10-06 19:09:38.000000000","message":"What happens with the volume data in this situation? Just lost and you\u0027v lost the volume? If you are overwriting the original volume and the operation fails this wouldn\u0027t  be acceptable.","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":132,"context_line":"    data from the rollback snapshot;"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":135,"context_line":"    is error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":136,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":137,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":138,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_62772ba0","line":135,"range":{"start_line":135,"start_character":14,"end_line":135,"end_character":33},"in_reply_to":"1a95cdbc_6bebd178","updated":"2016-10-09 02:54:43.000000000","message":"thanks, I add a first status check on volume and snapshot whether they are avaliable","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":131,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":132,"context_line":"    data from the rollback snapshot;"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":135,"context_line":"    is error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":136,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":137,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":138,"context_line":""}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_62d3ab60","line":135,"range":{"start_line":134,"start_character":4,"end_line":135,"end_character":14},"in_reply_to":"1a95cdbc_8bfbbdc0","updated":"2016-10-09 02:54:43.000000000","message":"in this case,because the error was met in the driver layer,what we can do is catch and store the error backtrace, change the status and then wait for the manual operation. so do you have any better solutions?","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":156,"context_line":""},{"line_number":157,"context_line":""},{"line_number":158,"context_line":"* If the cinder driver can not support the rollback snapshot feature,"},{"line_number":159,"context_line":"the framwork will use the general way to implement the rollback snapshot"},{"line_number":160,"context_line":"feature."},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"1. create a snapshot from volume"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_0e4fbb6b","line":159,"range":{"start_line":159,"start_character":4,"end_line":159,"end_character":12},"updated":"2016-10-06 19:09:38.000000000","message":"framework","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":156,"context_line":""},{"line_number":157,"context_line":""},{"line_number":158,"context_line":"* If the cinder driver can not support the rollback snapshot feature,"},{"line_number":159,"context_line":"the framwork will use the general way to implement the rollback snapshot"},{"line_number":160,"context_line":"feature."},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"1. create a snapshot from volume"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_8fc0e051","line":159,"range":{"start_line":159,"start_character":4,"end_line":159,"end_character":12},"in_reply_to":"1a95cdbc_0e4fbb6b","updated":"2016-10-09 02:54:43.000000000","message":"Done","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":167,"context_line":"If the cinder driver can not support mount snapshot, the general way"},{"line_number":168,"context_line":"will be the following steps."},{"line_number":169,"context_line":""},{"line_number":170,"context_line":"1. create a hidden volume"},{"line_number":171,"context_line":"2. mount hidden volume to host"},{"line_number":172,"context_line":"3. copy date from hidden volume to original volume"},{"line_number":173,"context_line":"4. delete the original volume"},{"line_number":174,"context_line":""},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"Alternatives"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_4efc6372","line":173,"range":{"start_line":170,"start_character":0,"end_line":173,"end_character":29},"updated":"2016-10-06 19:09:38.000000000","message":"To be as generic as possible, we need to use only this approach in all cases. Also, to avoid data lost, the sequence could be:\n\n0. create a snapshot (backup) of the target volume\n1. create a hidden volume from the target snapshot (the one we want to rollback)\n2. mount hidden volume to host\n3. copy data from hidden volume to original volume\n4. delete the backup snapshot\n5. delete the original volume","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":167,"context_line":"If the cinder driver can not support mount snapshot, the general way"},{"line_number":168,"context_line":"will be the following steps."},{"line_number":169,"context_line":""},{"line_number":170,"context_line":"1. create a hidden volume"},{"line_number":171,"context_line":"2. mount hidden volume to host"},{"line_number":172,"context_line":"3. copy date from hidden volume to original volume"},{"line_number":173,"context_line":"4. delete the original volume"},{"line_number":174,"context_line":""},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"Alternatives"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_6f8b941f","line":173,"range":{"start_line":170,"start_character":0,"end_line":173,"end_character":29},"in_reply_to":"1a95cdbc_4efc6372","updated":"2016-10-09 02:54:43.000000000","message":"thanks for your patient review,done","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"62ae10a6a5feef7eed9d58b84240bff5da3071a4","unresolved":false,"context_lines":[{"line_number":257,"context_line":"- Create driver interface"},{"line_number":258,"context_line":"- Implement LVM reference"},{"line_number":259,"context_line":"- Implement tempest test"},{"line_number":260,"context_line":"- Implement python-cinderclient"},{"line_number":261,"context_line":"- Implement Cinder UI"},{"line_number":262,"context_line":""},{"line_number":263,"context_line":"Dependencies"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_0e2dfbfd","line":260,"range":{"start_line":260,"start_character":0,"end_line":260,"end_character":31},"updated":"2016-10-06 19:09:38.000000000","message":"- Implement openstack-cinderclient","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"06c4a75b3e675f25d2479aa9b726afd8b555db81","unresolved":false,"context_lines":[{"line_number":257,"context_line":"- Create driver interface"},{"line_number":258,"context_line":"- Implement LVM reference"},{"line_number":259,"context_line":"- Implement tempest test"},{"line_number":260,"context_line":"- Implement python-cinderclient"},{"line_number":261,"context_line":"- Implement Cinder UI"},{"line_number":262,"context_line":""},{"line_number":263,"context_line":"Dependencies"}],"source_content_type":"text/x-rst","patch_set":14,"id":"1a95cdbc_ef7fa412","line":260,"range":{"start_line":260,"start_character":0,"end_line":260,"end_character":31},"in_reply_to":"1a95cdbc_0e2dfbfd","updated":"2016-10-09 02:54:43.000000000","message":"how about the usage of cinderclient to make it more clear and simple","commit_id":"773a63bb8c417cb071ce9ddaf105217343c4de5d"}],"specs/ocata/cinder-volume-revert-by-snapshot.rst":[{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"6fdecc65ffb3d76bd9cdc6867c505a74f6d29e3f","unresolved":false,"context_lines":[{"line_number":5,"context_line":" http://creativecommons.org/licenses/by/3.0/legalcode"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":8,"context_line":"Cinder volume revert from snapshot"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/cinder/+spec/revert-volume-to-snapshot"}],"source_content_type":"text/x-rst","patch_set":18,"id":"9a629dbe_38b993ea","line":8,"updated":"2016-11-07 15:50:29.000000000","message":"Isn\u0027t it revert _to_ snapshot?","commit_id":"165bfa65ded4ae71276222ff69ccf37eb3aa25b1"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"752b72d1e198b897d93996a3238960b4c5493ec1","unresolved":false,"context_lines":[{"line_number":5,"context_line":" http://creativecommons.org/licenses/by/3.0/legalcode"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":8,"context_line":"Cinder volume revert from snapshot"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/cinder/+spec/revert-volume-to-snapshot"}],"source_content_type":"text/x-rst","patch_set":18,"id":"9a629dbe_ce6d1506","line":8,"in_reply_to":"9a629dbe_38b993ea","updated":"2016-11-08 08:58:19.000000000","message":"+1","commit_id":"165bfa65ded4ae71276222ff69ccf37eb3aa25b1"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"9b54a0e032217665235da57d67b9602afdfa53e5","unresolved":false,"context_lines":[{"line_number":5,"context_line":" http://creativecommons.org/licenses/by/3.0/legalcode"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":8,"context_line":"Cinder volume revert from snapshot"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/cinder/+spec/revert-volume-to-snapshot"}],"source_content_type":"text/x-rst","patch_set":18,"id":"9a629dbe_ed37b6fc","line":8,"in_reply_to":"9a629dbe_38b993ea","updated":"2016-11-10 17:25:06.000000000","message":"+1","commit_id":"165bfa65ded4ae71276222ff69ccf37eb3aa25b1"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"6fdecc65ffb3d76bd9cdc6867c505a74f6d29e3f","unresolved":false,"context_lines":[{"line_number":152,"context_line":"feature."},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"1. create another hidden volume from snapshot"},{"line_number":155,"context_line":"2. swap the two volumes\u0027 identities(existing in migration logic)"},{"line_number":156,"context_line":"3. delete the hidden volume"},{"line_number":157,"context_line":""},{"line_number":158,"context_line":""}],"source_content_type":"text/x-rst","patch_set":18,"id":"9a629dbe_98037fb8","line":155,"updated":"2016-11-07 15:50:29.000000000","message":"This section needs more detail, I don\u0027t think it\u0027s quite this simple.\n\nYou would have to recreate the prior snapshots (before the one you are reverting to) as well on this new volume, or you lose all of your other snapshots.\n\nI\u0027m not sure that\u0027s feasible to implement.","commit_id":"165bfa65ded4ae71276222ff69ccf37eb3aa25b1"},{"author":{"_account_id":9003,"name":"Tom Barron","email":"tpb@dyncloud.net","username":"tbarron"},"change_message_id":"752b72d1e198b897d93996a3238960b4c5493ec1","unresolved":false,"context_lines":[{"line_number":152,"context_line":"feature."},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"1. create another hidden volume from snapshot"},{"line_number":155,"context_line":"2. swap the two volumes\u0027 identities(existing in migration logic)"},{"line_number":156,"context_line":"3. delete the hidden volume"},{"line_number":157,"context_line":""},{"line_number":158,"context_line":""}],"source_content_type":"text/x-rst","patch_set":18,"id":"9a629dbe_ae362101","line":155,"in_reply_to":"9a629dbe_98037fb8","updated":"2016-11-08 08:58:19.000000000","message":"Yeah, if there\u0027s a way around this issue it needs to be explained.","commit_id":"165bfa65ded4ae71276222ff69ccf37eb3aa25b1"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"9b54a0e032217665235da57d67b9602afdfa53e5","unresolved":false,"context_lines":[{"line_number":152,"context_line":"feature."},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"1. create another hidden volume from snapshot"},{"line_number":155,"context_line":"2. swap the two volumes\u0027 identities(existing in migration logic)"},{"line_number":156,"context_line":"3. delete the hidden volume"},{"line_number":157,"context_line":""},{"line_number":158,"context_line":""}],"source_content_type":"text/x-rst","patch_set":18,"id":"9a629dbe_8849c862","line":155,"in_reply_to":"9a629dbe_ae362101","updated":"2016-11-10 17:25:06.000000000","message":"There was a lot more information in previous PSs. What happened to then?","commit_id":"165bfa65ded4ae71276222ff69ccf37eb3aa25b1"}],"specs/ocata/cinder-volume-rollback-snapshot.rst":[{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":13,"context_line":""},{"line_number":14,"context_line":"The rollback function can be used to rollback a volume to a previous"},{"line_number":15,"context_line":"snapshot, restoring the volume to the state at the time the snapshot was"},{"line_number":16,"context_line":"taken. The function supports rollback to any previously taken snapshot on"},{"line_number":17,"context_line":"the volume(we will call it the rollback snapshot in this document to"},{"line_number":18,"context_line":"distinguish it from other snapshots)."},{"line_number":19,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_7516e64c","line":16,"range":{"start_line":16,"start_character":37,"end_line":16,"end_character":40},"updated":"2016-11-03 14:53:40.000000000","message":"To the most recent snapshot taken.\n* we need to explain somewhere (not in this intro) why this decision was made, which is, because some vendors does not support rolling back without loosing snapshots in the snapshot chain. I actually didn\u0027t quite understood who and how this problem came to happen. I\u0027ll try to clarify with Cinder folks.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":13,"context_line":""},{"line_number":14,"context_line":"The rollback function can be used to rollback a volume to a previous"},{"line_number":15,"context_line":"snapshot, restoring the volume to the state at the time the snapshot was"},{"line_number":16,"context_line":"taken. The function supports rollback to any previously taken snapshot on"},{"line_number":17,"context_line":"the volume(we will call it the rollback snapshot in this document to"},{"line_number":18,"context_line":"distinguish it from other snapshots)."},{"line_number":19,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_60e4143b","line":16,"range":{"start_line":16,"start_character":37,"end_line":16,"end_character":40},"in_reply_to":"ba5da102_7516e64c","updated":"2016-11-07 08:35:19.000000000","message":"Thanks, Erlon~","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":14,"context_line":"The rollback function can be used to rollback a volume to a previous"},{"line_number":15,"context_line":"snapshot, restoring the volume to the state at the time the snapshot was"},{"line_number":16,"context_line":"taken. The function supports rollback to any previously taken snapshot on"},{"line_number":17,"context_line":"the volume(we will call it the rollback snapshot in this document to"},{"line_number":18,"context_line":"distinguish it from other snapshots)."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"The rollback process will overwrite the existing volume with data from the"},{"line_number":21,"context_line":"rollback snapshot."}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_5559e233","line":18,"range":{"start_line":17,"start_character":10,"end_line":18,"end_character":37},"updated":"2016-11-03 14:53:40.000000000","message":"So, considering this situation, it won\u0027t be needed to specify which snapshot we are reverting to.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":14,"context_line":"The rollback function can be used to rollback a volume to a previous"},{"line_number":15,"context_line":"snapshot, restoring the volume to the state at the time the snapshot was"},{"line_number":16,"context_line":"taken. The function supports rollback to any previously taken snapshot on"},{"line_number":17,"context_line":"the volume(we will call it the rollback snapshot in this document to"},{"line_number":18,"context_line":"distinguish it from other snapshots)."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"The rollback process will overwrite the existing volume with data from the"},{"line_number":21,"context_line":"rollback snapshot."}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_a0d46c47","line":18,"range":{"start_line":17,"start_character":10,"end_line":18,"end_character":37},"in_reply_to":"ba5da102_5559e233","updated":"2016-11-07 08:35:19.000000000","message":"As Tom Barron metioned, more discussing still needed before we have a conclusion whether specify a snapshot id or not, his main concern is about openstack components\u0027 consistency.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":17,"context_line":"the volume(we will call it the rollback snapshot in this document to"},{"line_number":18,"context_line":"distinguish it from other snapshots)."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"The rollback process will overwrite the existing volume with data from the"},{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_754ee63b","line":21,"range":{"start_line":20,"start_character":40,"end_line":21,"end_character":18},"updated":"2016-11-03 14:53:40.000000000","message":"... the current state and data of the volume. If the volume was extended after the snapshot, it will return its previous size.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":17,"context_line":"the volume(we will call it the rollback snapshot in this document to"},{"line_number":18,"context_line":"distinguish it from other snapshots)."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"The rollback process will overwrite the existing volume with data from the"},{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_20da1c79","line":21,"range":{"start_line":20,"start_character":40,"end_line":21,"end_character":18},"in_reply_to":"ba5da102_754ee63b","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_b54ddec5","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"updated":"2016-11-03 14:53:40.000000000","message":"We should not make this a requirement. Reverting a volume without having to remove it is a great deal in the user POV and Nova does not allow to detach volumes that are boot/root disk. If the instance is shuttoff, we can safely revert its disk. As Cinder does not know the status of the instance, I can think in 2 options:\n 1 - Ask Nova to shutoff the instance (this is similar to vSphere) if the \u0027--force\u0027 is passed\n 2 - Give the user a \u0027--force\u0027 option that will warn the user that he/she must shutoff the instance and rely that he/she already did that. (This sound very dirty)","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"8eb123834a7dec6149b6ab68d7a3df5d82604333","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_eba3d9fe","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"in_reply_to":"9a629dbe_1bb27294","updated":"2016-11-13 15:02:24.000000000","message":"If we take the option where cinder does a shutdown of the instance itself, then there is still a locking problem between cinder and nova - there\u0027s nothing stopping the user (or, more likely, some other orchestration tool doing things automatically) from restarting the nova instance while cinder is still mid-way through the rollback.\n\nThere\u0027s also the problem that the shutdown can take a long time, which leaves cinder-api worker threads blocked, which can allow the cinder-api service to become unresponsive with enough of these going on - more so if there is a problem with the nova-api service. This can be reduced / alleviated by moving the shutdown into the c-vol service rather than doing it on the api. I\u0027ve done my best to resist blocking calling in c-api non-admin code paths for this reason.\n\nInter-service orchestration is hard. Really hard.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_00dd1870","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"in_reply_to":"ba5da102_0548d15a","updated":"2016-11-07 08:35:19.000000000","message":"@Duncan Thomas\nthe \u0027two API\u0027 here seems reasonable, but why should we expose a api called \u0027unsafe_rollback\u0027 to the inner components which ususally require a more \u0027safe\u0027 api?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"182795ee944efc411944983db941c5cf1f4930e2","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_1bb27294","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"in_reply_to":"ba5da102_0548d15a","updated":"2016-11-10 17:14:53.000000000","message":"@Duncan, in the option 1, which I think is the right path to go, there should be no data loss. The --force option is to let clear to the user that the instance the volume is attached will be shutdown. Even if we had any data loss here it wouldn\u0027t be a problem as it is expected to lost the current data of the volume.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"182795ee944efc411944983db941c5cf1f4930e2","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_019cbb0d","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"in_reply_to":"ba5da102_4062b0d9","updated":"2016-11-10 17:14:53.000000000","message":"Those kind of assumptions are too dangerous. If the user forget to shutdown the instance, the volume will be reverted and the living instance will be deadly broken trying to write things to files that no longer exists or are different from expected.\n\nThe shutdown can be done in Cinder without any nova change[1][2].\n\n[1] http://bit.ly/2eNENrM\n[2] http://bit.ly/2eNA4WU","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_4062b0d9","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"in_reply_to":"ba5da102_b54ddec5","updated":"2016-11-07 08:35:19.000000000","message":"@Erlon R. Cruz \nthis spec should focus on cinder and its related resource,\nmore automation(auto shutdown) would mean more dependencies on other openstack components and possible bugs here, if we can not gurantee the whole imaginary process, how about just leave this alone, let the end user ensure this ?:)","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":21,"context_line":"rollback snapshot."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"Note:"},{"line_number":24,"context_line":"The volume that for the rollback must be detached if it is attached to a"},{"line_number":25,"context_line":"Nova instance."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Problem description"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_0548d15a","line":25,"range":{"start_line":24,"start_character":0,"end_line":25,"end_character":14},"in_reply_to":"ba5da102_b54ddec5","updated":"2016-11-06 10:56:17.000000000","message":"Allowing a --force option in the API is, to my mind, a great mistake. It is an option the will, under nearly any circumstance except the one correct one, result in data corruption.\n\nI suspect the correct path forward here is to provide two APIs - one, the standard detached rollback, which would be plumbed though to the CLI, and a second (maybe called unsafe_rollback or something similarly scary sounding) that is intended to only be called by nova, similar to the attach and other nova-\u003ecinder APIs we already have.\n\nKeeping the two entirely separate makes for a much more obvious separation in their intended usages.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":54,"context_line":"  rollback snapshot, it will keep the same volume id"},{"line_number":55,"context_line":"  (and therefore disk UUID) and makes reverting a instance much easier since"},{"line_number":56,"context_line":"  device names inside the instance stay the same."},{"line_number":57,"context_line":""},{"line_number":58,"context_line":"* Use restore backup, restore backup data to volume may be a way to recover"},{"line_number":59,"context_line":"  data when the storage does not exist, but most time the instance can\u0027t work"},{"line_number":60,"context_line":"  may be not the storage\u0027s problem, the problem may be only some data is not"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_c03372f4","line":57,"updated":"2016-11-03 14:53:40.000000000","message":"* After creating a new instance from the volume, the instance receives new IP address which can lead to several problems when other applications rely in that host IP.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":54,"context_line":"  rollback snapshot, it will keep the same volume id"},{"line_number":55,"context_line":"  (and therefore disk UUID) and makes reverting a instance much easier since"},{"line_number":56,"context_line":"  device names inside the instance stay the same."},{"line_number":57,"context_line":""},{"line_number":58,"context_line":"* Use restore backup, restore backup data to volume may be a way to recover"},{"line_number":59,"context_line":"  data when the storage does not exist, but most time the instance can\u0027t work"},{"line_number":60,"context_line":"  may be not the storage\u0027s problem, the problem may be only some data is not"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_d58760d5","line":57,"in_reply_to":"ba5da102_c03372f4","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":55,"context_line":"  (and therefore disk UUID) and makes reverting a instance much easier since"},{"line_number":56,"context_line":"  device names inside the instance stay the same."},{"line_number":57,"context_line":""},{"line_number":58,"context_line":"* Use restore backup, restore backup data to volume may be a way to recover"},{"line_number":59,"context_line":"  data when the storage does not exist, but most time the instance can\u0027t work"},{"line_number":60,"context_line":"  may be not the storage\u0027s problem, the problem may be only some data is not"},{"line_number":61,"context_line":"  concurrent with the system. So we also hope that the data can recover"},{"line_number":62,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"},{"line_number":63,"context_line":"  find it may spend more time to copy the data to the new volume."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":""},{"line_number":66,"context_line":"As mentioned above, we hope that can offer a way to recover instance quickly"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_a5956589","line":63,"range":{"start_line":58,"start_character":2,"end_line":63,"end_character":65},"updated":"2016-11-06 10:56:17.000000000","message":"I\u0027m having trouble understanding exactly what you\u0027re saying here. I *think* you\u0027re saying that restore from backup to the existing volume could be used here, but the performance is poor? If so, I agree.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":55,"context_line":"  (and therefore disk UUID) and makes reverting a instance much easier since"},{"line_number":56,"context_line":"  device names inside the instance stay the same."},{"line_number":57,"context_line":""},{"line_number":58,"context_line":"* Use restore backup, restore backup data to volume may be a way to recover"},{"line_number":59,"context_line":"  data when the storage does not exist, but most time the instance can\u0027t work"},{"line_number":60,"context_line":"  may be not the storage\u0027s problem, the problem may be only some data is not"},{"line_number":61,"context_line":"  concurrent with the system. So we also hope that the data can recover"},{"line_number":62,"context_line":"  from local, not the remote devices. If we use restore backup, we may also"},{"line_number":63,"context_line":"  find it may spend more time to copy the data to the new volume."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":""},{"line_number":66,"context_line":"As mentioned above, we hope that can offer a way to recover instance quickly"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_9564084a","line":63,"range":{"start_line":58,"start_character":2,"end_line":63,"end_character":65},"in_reply_to":"ba5da102_a5956589","updated":"2016-11-07 08:35:19.000000000","message":"yeah, you got it. I reorgnised this paragraph, wish it works~","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":69,"context_line":"Use Cases"},{"line_number":70,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"This is intended to be a DR function. The primary use case is for a instance"},{"line_number":73,"context_line":"fall down but the storage is also live, so rollback snapshot to the volume"},{"line_number":74,"context_line":"main user case."},{"line_number":75,"context_line":"1. User creates a volume(s)"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_c01ff2b4","line":72,"range":{"start_line":72,"start_character":25,"end_line":72,"end_character":27},"updated":"2016-11-03 14:53:40.000000000","message":"DR?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":69,"context_line":"Use Cases"},{"line_number":70,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"This is intended to be a DR function. The primary use case is for a instance"},{"line_number":73,"context_line":"fall down but the storage is also live, so rollback snapshot to the volume"},{"line_number":74,"context_line":"main user case."},{"line_number":75,"context_line":"1. User creates a volume(s)"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_f5daa4e8","line":72,"range":{"start_line":72,"start_character":25,"end_line":72,"end_character":27},"in_reply_to":"ba5da102_c01ff2b4","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":76,"context_line":"2. User attach volume to instance"},{"line_number":77,"context_line":"3. User creates a snapshot"},{"line_number":78,"context_line":"4. Something happens causing the need of a rollback"},{"line_number":79,"context_line":"5. User detach volume from instance"},{"line_number":80,"context_line":"6. User rollback snapshot(s)"},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"We already have create a new volume from snapshot function, Why do we need"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_e0f6b65b","line":79,"range":{"start_line":79,"start_character":3,"end_line":79,"end_character":35},"updated":"2016-11-03 14:53:40.000000000","message":"Shutdown the intance","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"182795ee944efc411944983db941c5cf1f4930e2","unresolved":false,"context_lines":[{"line_number":76,"context_line":"2. User attach volume to instance"},{"line_number":77,"context_line":"3. User creates a snapshot"},{"line_number":78,"context_line":"4. Something happens causing the need of a rollback"},{"line_number":79,"context_line":"5. User detach volume from instance"},{"line_number":80,"context_line":"6. User rollback snapshot(s)"},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"We already have create a new volume from snapshot function, Why do we need"}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_92ef37e8","line":79,"range":{"start_line":79,"start_character":3,"end_line":79,"end_character":35},"in_reply_to":"ba5da102_5555101b","updated":"2016-11-10 17:14:53.000000000","message":"If I understood correctly, what Duncan is proposing is what I mentioned above about calling nova.API() to shutdown or block operations in the instance.\n\nIf we only support the rollback of attached snapshots, we loose most of the use cases useful to the revert feature as is not possible detach bootable volumes.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":76,"context_line":"2. User attach volume to instance"},{"line_number":77,"context_line":"3. User creates a snapshot"},{"line_number":78,"context_line":"4. Something happens causing the need of a rollback"},{"line_number":79,"context_line":"5. User detach volume from instance"},{"line_number":80,"context_line":"6. User rollback snapshot(s)"},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"We already have create a new volume from snapshot function, Why do we need"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_5555101b","line":79,"range":{"start_line":79,"start_character":3,"end_line":79,"end_character":35},"in_reply_to":"ba5da102_85f781f4","updated":"2016-11-07 08:35:19.000000000","message":"@Erlon R. Cruz and @Duncan Thomas,\nAgreed with Duncan Thomas, I think it\u0027s better to keep the preset and process simple, which means the only thing we need to gurantee before the reverting is the \u0027avaliable\u0027 status of volume, the \u0027shutdown, dettach\u0027 or anything could make the status sastified is repsonsible by user, also we can use \u0027reverting\u0027 status for preventing incoming writing operations.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":76,"context_line":"2. User attach volume to instance"},{"line_number":77,"context_line":"3. User creates a snapshot"},{"line_number":78,"context_line":"4. Something happens causing the need of a rollback"},{"line_number":79,"context_line":"5. User detach volume from instance"},{"line_number":80,"context_line":"6. User rollback snapshot(s)"},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"We already have create a new volume from snapshot function, Why do we need"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_85f781f4","line":79,"range":{"start_line":79,"start_character":3,"end_line":79,"end_character":35},"in_reply_to":"ba5da102_e0f6b65b","updated":"2016-11-06 10:56:17.000000000","message":"Any workflow requiring the instance to be shut down to avoid data loss/corruption should be orchestrated by nova, not done via cinder. Or orchestrated by cinder via some (probably new) cinder-\u003enova API that ensures the instance can\u0027t be started until the rollback is done. Avoiding data corruption should not be left to the tenant to orchestrate the API calls correctly.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":97,"context_line":""},{"line_number":98,"context_line":"Proposed change"},{"line_number":99,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":100,"context_line":"Theoretically, if multiple snapshots exist, a storage system could restore"},{"line_number":101,"context_line":"a volume to any one of those snapshots. But rolling back a volume to a"},{"line_number":102,"context_line":"snapshot which is not the most recent one typically deletes the later"},{"line_number":103,"context_line":"snapshots, which is a form of data loss and may not be what the user intends."},{"line_number":104,"context_line":"So this feature will limit cinder to rolling back the most recent volume"},{"line_number":105,"context_line":"snapshot known to cinder."},{"line_number":106,"context_line":""},{"line_number":107,"context_line":"By supporting rollback snapshot, we need to do the following changes."},{"line_number":108,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_25cf95b4","line":105,"range":{"start_line":100,"start_character":0,"end_line":105,"end_character":25},"updated":"2016-11-06 10:56:17.000000000","message":"We should add a caveat that rolling back to a snapshot does *not* delete the snapshot. Some vendor(s) were talking about that behavior, I think the implicit delete is a violation of expectations of the cinder API.\n\nThis means that the following workflow is valid:\n\n\u003ccreate snap 1\u003e\n\u003cdo work on the instance\u003e\n\u003cdiscover problem\u003e\n\u003croll back to snap 1\u003e\n\u003ctry again\u003e\n\u003cdiscover problem\u003e\n\u003croll back to snap 1\u003e","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":97,"context_line":""},{"line_number":98,"context_line":"Proposed change"},{"line_number":99,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":100,"context_line":"Theoretically, if multiple snapshots exist, a storage system could restore"},{"line_number":101,"context_line":"a volume to any one of those snapshots. But rolling back a volume to a"},{"line_number":102,"context_line":"snapshot which is not the most recent one typically deletes the later"},{"line_number":103,"context_line":"snapshots, which is a form of data loss and may not be what the user intends."},{"line_number":104,"context_line":"So this feature will limit cinder to rolling back the most recent volume"},{"line_number":105,"context_line":"snapshot known to cinder."},{"line_number":106,"context_line":""},{"line_number":107,"context_line":"By supporting rollback snapshot, we need to do the following changes."},{"line_number":108,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_f0daafe8","line":105,"range":{"start_line":100,"start_character":0,"end_line":105,"end_character":25},"in_reply_to":"ba5da102_25cf95b4","updated":"2016-11-07 08:35:19.000000000","message":"I do not quite understand, you mean we should explicit point out the latest snapshot won\u0027t be deleted by reverting action and can be used many times?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"182795ee944efc411944983db941c5cf1f4930e2","unresolved":false,"context_lines":[{"line_number":97,"context_line":""},{"line_number":98,"context_line":"Proposed change"},{"line_number":99,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":100,"context_line":"Theoretically, if multiple snapshots exist, a storage system could restore"},{"line_number":101,"context_line":"a volume to any one of those snapshots. But rolling back a volume to a"},{"line_number":102,"context_line":"snapshot which is not the most recent one typically deletes the later"},{"line_number":103,"context_line":"snapshots, which is a form of data loss and may not be what the user intends."},{"line_number":104,"context_line":"So this feature will limit cinder to rolling back the most recent volume"},{"line_number":105,"context_line":"snapshot known to cinder."},{"line_number":106,"context_line":""},{"line_number":107,"context_line":"By supporting rollback snapshot, we need to do the following changes."},{"line_number":108,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_d22c4f04","line":105,"range":{"start_line":100,"start_character":0,"end_line":105,"end_character":25},"in_reply_to":"ba5da102_f0daafe8","updated":"2016-11-10 17:14:53.000000000","message":"I think that the text can give the impression that the revert snapshot will delete the snapshot we are reverting to (which are not what we are considering). See the \u0027proposed\u0027 change\u0027 section in manila[1].\n\n[1] https://review.openstack.org/#/c/391269/","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":106,"context_line":""},{"line_number":107,"context_line":"By supporting rollback snapshot, we need to do the following changes."},{"line_number":108,"context_line":""},{"line_number":109,"context_line":"Add rollback snapshot API."},{"line_number":110,"context_line":""},{"line_number":111,"context_line":"::"},{"line_number":112,"context_line":"    cinder snapshot-rollback \u003cvolume\u003e"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_fbebc710","line":109,"range":{"start_line":109,"start_character":0,"end_line":109,"end_character":3},"updated":"2016-11-03 14:53:40.000000000","message":"* Add","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":106,"context_line":""},{"line_number":107,"context_line":"By supporting rollback snapshot, we need to do the following changes."},{"line_number":108,"context_line":""},{"line_number":109,"context_line":"Add rollback snapshot API."},{"line_number":110,"context_line":""},{"line_number":111,"context_line":"::"},{"line_number":112,"context_line":"    cinder snapshot-rollback \u003cvolume\u003e"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_cdafd892","line":109,"range":{"start_line":109,"start_character":0,"end_line":109,"end_character":3},"in_reply_to":"ba5da102_fbebc710","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":123,"context_line":""},{"line_number":124,"context_line":"    If the cinder driver can not support the rollback feature, it will"},{"line_number":125,"context_line":"    use the generic implementation to run rollback feature."},{"line_number":126,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":127,"context_line":"    data from the rollback snapshot;"},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_c0053214","line":126,"range":{"start_line":126,"start_character":30,"end_line":126,"end_character":39},"updated":"2016-11-03 14:53:40.000000000","message":"discard the current volume data","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":123,"context_line":""},{"line_number":124,"context_line":"    If the cinder driver can not support the rollback feature, it will"},{"line_number":125,"context_line":"    use the generic implementation to run rollback feature."},{"line_number":126,"context_line":"    The rollback process will overwrite the existing volume with"},{"line_number":127,"context_line":"    data from the rollback snapshot;"},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_6d2a4ce9","line":126,"range":{"start_line":126,"start_character":30,"end_line":126,"end_character":39},"in_reply_to":"ba5da102_c0053214","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":127,"context_line":"    data from the rollback snapshot;"},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":130,"context_line":"    is error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":131,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":132,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":133,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_1b2cfb97","line":130,"range":{"start_line":130,"start_character":24,"end_line":130,"end_character":33},"updated":"2016-11-03 14:53:40.000000000","message":"instance is on","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":127,"context_line":"    data from the rollback snapshot;"},{"line_number":128,"context_line":""},{"line_number":129,"context_line":"    For lvm driver, if rollback snapshot fail, the volume\u0027s status"},{"line_number":130,"context_line":"    is error. If volume is in use, it will raise an VolumeIsBusy"},{"line_number":131,"context_line":"    exception to volume manager, the volume\u0027s status is still available,"},{"line_number":132,"context_line":"    because the original volume data has not been changed at this time."},{"line_number":133,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_8d2fa0f5","line":130,"range":{"start_line":130,"start_character":24,"end_line":130,"end_character":33},"in_reply_to":"ba5da102_1b2cfb97","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":134,"context_line":"* Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":135,"context_line":"driver."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"Vendors can implement the rollback snapshot feature in their drivers in"},{"line_number":138,"context_line":"order to rollback snapshot to volume without requiring create a new volume"},{"line_number":139,"context_line":"from snapshot."},{"line_number":140,"context_line":""},{"line_number":141,"context_line":"Add driver interfaces::"},{"line_number":142,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_fb9d87bc","line":139,"range":{"start_line":137,"start_character":0,"end_line":139,"end_character":13},"updated":"2016-11-03 14:53:40.000000000","message":"* Add a function that vendors can use to override the default rollback_snapshot function with an optimized version.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":134,"context_line":"* Add rollback_snapshot in volume driver and implement this interface in LVM"},{"line_number":135,"context_line":"driver."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"Vendors can implement the rollback snapshot feature in their drivers in"},{"line_number":138,"context_line":"order to rollback snapshot to volume without requiring create a new volume"},{"line_number":139,"context_line":"from snapshot."},{"line_number":140,"context_line":""},{"line_number":141,"context_line":"Add driver interfaces::"},{"line_number":142,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_8d16c0a5","line":139,"range":{"start_line":137,"start_character":0,"end_line":139,"end_character":13},"in_reply_to":"ba5da102_fb9d87bc","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":147,"context_line":"        :param volume: the volume to be rollbacked."},{"line_number":148,"context_line":"        :param snapshot: the snapshot data rollback to volume."},{"line_number":149,"context_line":"        \"\"\""},{"line_number":150,"context_line":""},{"line_number":151,"context_line":""},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"* If the cinder driver can not support the rollback snapshot feature,"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_9b6bab1d","line":150,"updated":"2016-11-03 14:53:40.000000000","message":"What does the driver needs to return to the caller?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":147,"context_line":"        :param volume: the volume to be rollbacked."},{"line_number":148,"context_line":"        :param snapshot: the snapshot data rollback to volume."},{"line_number":149,"context_line":"        \"\"\""},{"line_number":150,"context_line":""},{"line_number":151,"context_line":""},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"* If the cinder driver can not support the rollback snapshot feature,"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_8d67a024","line":150,"in_reply_to":"ba5da102_9b6bab1d","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":155,"context_line":"feature."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"1. create a snapshot from volume"},{"line_number":158,"context_line":"2. mount snapshot to host"},{"line_number":159,"context_line":"3. copy date from snapshot to original volume"},{"line_number":160,"context_line":"4. delete the original volume"},{"line_number":161,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_bb2eafa2","line":158,"range":{"start_line":158,"start_character":3,"end_line":158,"end_character":17},"updated":"2016-11-03 14:53:40.000000000","message":"We shouldn\u0027t rely on mounting a snapshot. Cinder does not have support to it, and most of backends also dont.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"182795ee944efc411944983db941c5cf1f4930e2","unresolved":false,"context_lines":[{"line_number":156,"context_line":""},{"line_number":157,"context_line":"1. create a snapshot from volume"},{"line_number":158,"context_line":"2. mount snapshot to host"},{"line_number":159,"context_line":"3. copy date from snapshot to original volume"},{"line_number":160,"context_line":"4. delete the original volume"},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"If the cinder driver can not support mount snapshot, the general way"}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_d4efdf84","line":159,"range":{"start_line":159,"start_character":8,"end_line":159,"end_character":12},"updated":"2016-11-10 17:14:53.000000000","message":"data","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":165,"context_line":"1. create a snapshot from volume"},{"line_number":166,"context_line":"2. create a hidden volume from snapshot"},{"line_number":167,"context_line":"3. mount hidden volume to host"},{"line_number":168,"context_line":"4. copy date from hidden volume to original volume"},{"line_number":169,"context_line":"5. delete the backup snapshot"},{"line_number":170,"context_line":"6. delete the original volume"},{"line_number":171,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_e5bd9d03","line":168,"range":{"start_line":168,"start_character":0,"end_line":168,"end_character":50},"updated":"2016-11-06 10:56:17.000000000","message":"No need to copy data, we can use the volume id mangling framework from the migration code to change the volume id of the new volume in the DB. The generic implementation is as simple as create volume from snap, swap volume ids, delete original volume.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"182795ee944efc411944983db941c5cf1f4930e2","unresolved":false,"context_lines":[{"line_number":165,"context_line":"1. create a snapshot from volume"},{"line_number":166,"context_line":"2. create a hidden volume from snapshot"},{"line_number":167,"context_line":"3. mount hidden volume to host"},{"line_number":168,"context_line":"4. copy date from hidden volume to original volume"},{"line_number":169,"context_line":"5. delete the backup snapshot"},{"line_number":170,"context_line":"6. delete the original volume"},{"line_number":171,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"9a629dbe_6d7ba628","line":168,"range":{"start_line":168,"start_character":0,"end_line":168,"end_character":50},"in_reply_to":"ba5da102_e3958b88","updated":"2016-11-10 17:14:53.000000000","message":"@Duncan, as we are considering to make possible to revert attached snapshots, the swapping process might not work. Do you think it possible?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":165,"context_line":"1. create a snapshot from volume"},{"line_number":166,"context_line":"2. create a hidden volume from snapshot"},{"line_number":167,"context_line":"3. mount hidden volume to host"},{"line_number":168,"context_line":"4. copy date from hidden volume to original volume"},{"line_number":169,"context_line":"5. delete the backup snapshot"},{"line_number":170,"context_line":"6. delete the original volume"},{"line_number":171,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_e3958b88","line":168,"range":{"start_line":168,"start_character":0,"end_line":168,"end_character":50},"in_reply_to":"ba5da102_e5bd9d03","updated":"2016-11-07 08:35:19.000000000","message":"I got your point, but still need further study on this mangling framework, would you please paste a spec link here?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":167,"context_line":"3. mount hidden volume to host"},{"line_number":168,"context_line":"4. copy date from hidden volume to original volume"},{"line_number":169,"context_line":"5. delete the backup snapshot"},{"line_number":170,"context_line":"6. delete the original volume"},{"line_number":171,"context_line":""},{"line_number":172,"context_line":""},{"line_number":173,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_9b23abbc","line":170,"range":{"start_line":170,"start_character":14,"end_line":170,"end_character":22},"updated":"2016-11-03 14:53:40.000000000","message":"6. delete the hidden volume\nI said \u0027original\u0027 in my last review but the correct is the hidden volume. We should not delete the volume we are working on :), sorry.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":167,"context_line":"3. mount hidden volume to host"},{"line_number":168,"context_line":"4. copy date from hidden volume to original volume"},{"line_number":169,"context_line":"5. delete the backup snapshot"},{"line_number":170,"context_line":"6. delete the original volume"},{"line_number":171,"context_line":""},{"line_number":172,"context_line":""},{"line_number":173,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_0d09d031","line":170,"range":{"start_line":170,"start_character":14,"end_line":170,"end_character":22},"in_reply_to":"ba5da102_9b23abbc","updated":"2016-11-07 08:35:19.000000000","message":"thanks~","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":174,"context_line":"Alternatives"},{"line_number":175,"context_line":"------------"},{"line_number":176,"context_line":""},{"line_number":177,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":178,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Data model impact"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_0629028a","line":177,"range":{"start_line":177,"start_character":0,"end_line":177,"end_character":3},"updated":"2016-11-03 14:53:40.000000000","message":"* The ...","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":174,"context_line":"Alternatives"},{"line_number":175,"context_line":"------------"},{"line_number":176,"context_line":""},{"line_number":177,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":178,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Data model impact"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_f039cfd9","line":177,"range":{"start_line":177,"start_character":0,"end_line":177,"end_character":3},"in_reply_to":"ba5da102_0629028a","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":175,"context_line":"------------"},{"line_number":176,"context_line":""},{"line_number":177,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":178,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Data model impact"},{"line_number":181,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_8616f249","line":178,"range":{"start_line":178,"start_character":49,"end_line":178,"end_character":73},"updated":"2016-11-03 14:53:40.000000000","message":"... this alternative has the drawbacks described in the \u0027Problem Description\u0027 section.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":175,"context_line":"------------"},{"line_number":176,"context_line":""},{"line_number":177,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":178,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Data model impact"},{"line_number":181,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_50319beb","line":178,"range":{"start_line":178,"start_character":49,"end_line":178,"end_character":73},"in_reply_to":"ba5da102_8616f249","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":176,"context_line":""},{"line_number":177,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":178,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Data model impact"},{"line_number":181,"context_line":"-----------------"},{"line_number":182,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_66a71e0b","line":179,"updated":"2016-11-03 14:53:40.000000000","message":"* Implement snapshot revert allowing revert to any snapshot the volume has.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":176,"context_line":""},{"line_number":177,"context_line":"The alternative is to create a new volume from the snapshot. While this is"},{"line_number":178,"context_line":"already implemented and usable, this alternative consumes the user quotas."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"Data model impact"},{"line_number":181,"context_line":"-----------------"},{"line_number":182,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_d0720bad","line":179,"in_reply_to":"ba5da102_66a71e0b","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":180,"context_line":"Data model impact"},{"line_number":181,"context_line":"-----------------"},{"line_number":182,"context_line":""},{"line_number":183,"context_line":"None"},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"REST API impact"},{"line_number":186,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_255dd5ec","line":183,"range":{"start_line":183,"start_character":0,"end_line":183,"end_character":4},"updated":"2016-11-06 10:56:17.000000000","message":"We might want to introduce a \u0027rolling-back\u0027 state to volumes, to make locking easier (e.g. block an incoming attach against an ongoing rollback request)","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":180,"context_line":"Data model impact"},{"line_number":181,"context_line":"-----------------"},{"line_number":182,"context_line":""},{"line_number":183,"context_line":"None"},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"REST API impact"},{"line_number":186,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_b0a4b708","line":183,"range":{"start_line":183,"start_character":0,"end_line":183,"end_character":4},"in_reply_to":"ba5da102_255dd5ec","updated":"2016-11-07 08:35:19.000000000","message":"cool~, thanks","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":191,"context_line":"* Rollback snapshot"},{"line_number":192,"context_line":"* ``/v3/{tenant_id}/volumes/{volume_id}/rollback``"},{"line_number":193,"context_line":"* Method: POST"},{"line_number":194,"context_line":"* Request Body is not required"},{"line_number":195,"context_line":"Security impact"},{"line_number":196,"context_line":"---------------"},{"line_number":197,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_266446b7","line":194,"updated":"2016-11-03 14:53:40.000000000","message":"What is the response expected?","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":191,"context_line":"* Rollback snapshot"},{"line_number":192,"context_line":"* ``/v3/{tenant_id}/volumes/{volume_id}/rollback``"},{"line_number":193,"context_line":"* Method: POST"},{"line_number":194,"context_line":"* Request Body is not required"},{"line_number":195,"context_line":"Security impact"},{"line_number":196,"context_line":"---------------"},{"line_number":197,"context_line":""}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_3090a795","line":194,"in_reply_to":"ba5da102_266446b7","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":200,"context_line":"Notifications impact"},{"line_number":201,"context_line":"--------------------"},{"line_number":202,"context_line":""},{"line_number":203,"context_line":"None"},{"line_number":204,"context_line":""},{"line_number":205,"context_line":"Other end user impact"},{"line_number":206,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_855ce1ee","line":203,"range":{"start_line":203,"start_character":0,"end_line":203,"end_character":4},"updated":"2016-11-06 10:56:17.000000000","message":"We should absolutely emit a notification when this action happens, including details like whether the size changed due to this call.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":200,"context_line":"Notifications impact"},{"line_number":201,"context_line":"--------------------"},{"line_number":202,"context_line":""},{"line_number":203,"context_line":"None"},{"line_number":204,"context_line":""},{"line_number":205,"context_line":"Other end user impact"},{"line_number":206,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_c3396743","line":203,"range":{"start_line":203,"start_character":0,"end_line":203,"end_character":4},"in_reply_to":"ba5da102_855ce1ee","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":210,"context_line":"Performance Impact"},{"line_number":211,"context_line":"------------------"},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"The existing create-volume-from-snapshot workflow must be protected with"},{"line_number":214,"context_line":"locks to ensure no races occur in c-vol."},{"line_number":215,"context_line":""},{"line_number":216,"context_line":""},{"line_number":217,"context_line":"Other deployer impact"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_86d19225","line":214,"range":{"start_line":213,"start_character":0,"end_line":214,"end_character":40},"updated":"2016-11-03 14:53:40.000000000","message":"This is not a performance impact. Also, I think this is already done in Cinder.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"68dfc4b16353e4d9bca1fe321aeee9eee61b74aa","unresolved":false,"context_lines":[{"line_number":212,"context_line":""},{"line_number":213,"context_line":"The existing create-volume-from-snapshot workflow must be protected with"},{"line_number":214,"context_line":"locks to ensure no races occur in c-vol."},{"line_number":215,"context_line":""},{"line_number":216,"context_line":""},{"line_number":217,"context_line":"Other deployer impact"},{"line_number":218,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_26cd2631","line":215,"updated":"2016-11-03 14:53:40.000000000","message":"The proposed working flow might rely in 3 copy operations:\n1 - Take an snapshot of the target volume\n2 - Create a volume from the target snapshot\n3 - Copy the data from the hidden volume to the target volume\n\nThese 3 copy operations can take a long time, depending on the size of the volume, and therefore it can lock the storage for long periods of times.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":212,"context_line":""},{"line_number":213,"context_line":"The existing create-volume-from-snapshot workflow must be protected with"},{"line_number":214,"context_line":"locks to ensure no races occur in c-vol."},{"line_number":215,"context_line":""},{"line_number":216,"context_line":""},{"line_number":217,"context_line":"Other deployer impact"},{"line_number":218,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_a571e579","line":215,"in_reply_to":"ba5da102_26cd2631","updated":"2016-11-06 10:56:17.000000000","message":"No data copy is needed. See above.","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":212,"context_line":""},{"line_number":213,"context_line":"The existing create-volume-from-snapshot workflow must be protected with"},{"line_number":214,"context_line":"locks to ensure no races occur in c-vol."},{"line_number":215,"context_line":""},{"line_number":216,"context_line":""},{"line_number":217,"context_line":"Other deployer impact"},{"line_number":218,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_434b97b0","line":215,"in_reply_to":"ba5da102_a571e579","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":222,"context_line":"Developer impact"},{"line_number":223,"context_line":"----------------"},{"line_number":224,"context_line":""},{"line_number":225,"context_line":"Driver maintainers can add the new calls to bring this functionality to"},{"line_number":226,"context_line":"their drivers."},{"line_number":227,"context_line":""},{"line_number":228,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_c56ed995","line":225,"range":{"start_line":225,"start_character":44,"end_line":225,"end_character":54},"updated":"2016-11-06 10:56:17.000000000","message":"optimised versions of this functionality","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":222,"context_line":"Developer impact"},{"line_number":223,"context_line":"----------------"},{"line_number":224,"context_line":""},{"line_number":225,"context_line":"Driver maintainers can add the new calls to bring this functionality to"},{"line_number":226,"context_line":"their drivers."},{"line_number":227,"context_line":""},{"line_number":228,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_903b339c","line":225,"range":{"start_line":225,"start_character":44,"end_line":225,"end_character":54},"in_reply_to":"ba5da102_c56ed995","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":1207,"name":"Duncan Thomas","email":"duncan.thomas@gmail.com","username":"duncan-thomas"},"change_message_id":"b6eda1f099995cfd9fbdec0dcfe3c57958ee9954","unresolved":false,"context_lines":[{"line_number":240,"context_line":""},{"line_number":241,"context_line":"- Create APIs"},{"line_number":242,"context_line":"- Create RPC calls"},{"line_number":243,"context_line":"- Create driver interface"},{"line_number":244,"context_line":"- Implement LVM reference"},{"line_number":245,"context_line":"- Implement tempest test"},{"line_number":246,"context_line":"- Implement Cinder Client"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_e56bdd83","line":243,"range":{"start_line":243,"start_character":0,"end_line":243,"end_character":25},"updated":"2016-11-06 10:56:17.000000000","message":"The generic implementation should be merged before the optimised driver stuff, not after","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"84a3c40f8a96ff25f5db1e62f8fe405aaf747f43","unresolved":false,"context_lines":[{"line_number":240,"context_line":""},{"line_number":241,"context_line":"- Create APIs"},{"line_number":242,"context_line":"- Create RPC calls"},{"line_number":243,"context_line":"- Create driver interface"},{"line_number":244,"context_line":"- Implement LVM reference"},{"line_number":245,"context_line":"- Implement tempest test"},{"line_number":246,"context_line":"- Implement Cinder Client"}],"source_content_type":"text/x-rst","patch_set":16,"id":"ba5da102_900973e0","line":243,"range":{"start_line":243,"start_character":0,"end_line":243,"end_character":25},"in_reply_to":"ba5da102_e56bdd83","updated":"2016-11-07 08:35:19.000000000","message":"Done","commit_id":"54198fb7040ec4fd2ad82decd23b6bc45e5b529c"}],"specs/pike/cinder-volume-revert-by-snapshot.rst":[{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"882abe724abee49c1f0e827afcde4a4e538cc237","unresolved":false,"context_lines":[{"line_number":1,"context_line":".."},{"line_number":2,"context_line":" This work is licensed under a Creative Commons Attribution 3.0 Unported"},{"line_number":3,"context_line":" License."},{"line_number":4,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"7a77a97e_3e5a312b","line":1,"range":{"start_line":1,"start_character":0,"end_line":1,"end_character":2},"updated":"2016-11-22 18:21:21.000000000","message":"Hmmm, we\u0027ve lost the history (diff is empty), but its ok. Next time when you need to move folders in git, first move, post to gerrit, and then add your changes. Then you keep the diffs.","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2e030971ad8e247060050680eb2dedc8f8f7fd9d","unresolved":false,"context_lines":[{"line_number":1,"context_line":".."},{"line_number":2,"context_line":" This work is licensed under a Creative Commons Attribution 3.0 Unported"},{"line_number":3,"context_line":" License."},{"line_number":4,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_0ad647c5","line":1,"range":{"start_line":1,"start_character":0,"end_line":1,"end_character":2},"in_reply_to":"7a77a97e_3e5a312b","updated":"2016-11-23 03:25:05.000000000","message":"Thanks, got it~","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"882abe724abee49c1f0e827afcde4a4e538cc237","unresolved":false,"context_lines":[{"line_number":156,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":157,"context_line":"    \u0027reverting\u0027, the snapshot\u0027s status is also \u0027restoring\u0027. Other"},{"line_number":158,"context_line":"    operations must not be allowed to either object during the"},{"line_number":159,"context_line":"    restoration (Nova with new changes will block incoming operations with"},{"line_number":160,"context_line":"    volume status of \u0027reverting\u0027)."},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"    If the cinder driver can not support the revert feature, it will"},{"line_number":163,"context_line":"    use the generic implementation to run revert process."}],"source_content_type":"text/x-rst","patch_set":19,"id":"7a77a97e_e991c847","line":160,"range":{"start_line":159,"start_character":16,"end_line":160,"end_character":33},"updated":"2016-11-22 18:21:21.000000000","message":"Didn\u0027t understood this part and how Nova would be involved here.","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2e030971ad8e247060050680eb2dedc8f8f7fd9d","unresolved":false,"context_lines":[{"line_number":156,"context_line":"    affected whose states must be reflected. The volume\u0027s status is"},{"line_number":157,"context_line":"    \u0027reverting\u0027, the snapshot\u0027s status is also \u0027restoring\u0027. Other"},{"line_number":158,"context_line":"    operations must not be allowed to either object during the"},{"line_number":159,"context_line":"    restoration (Nova with new changes will block incoming operations with"},{"line_number":160,"context_line":"    volume status of \u0027reverting\u0027)."},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"    If the cinder driver can not support the revert feature, it will"},{"line_number":163,"context_line":"    use the generic implementation to run revert process."}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_3edd0871","line":160,"range":{"start_line":159,"start_character":16,"end_line":160,"end_character":33},"in_reply_to":"7a77a97e_e991c847","updated":"2016-11-23 03:25:05.000000000","message":"Sorry, reorganised my words","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"882abe724abee49c1f0e827afcde4a4e538cc237","unresolved":false,"context_lines":[{"line_number":243,"context_line":"Notifications about the revert process(start, end, error) and"},{"line_number":244,"context_line":"so does the related change notification will be add."},{"line_number":245,"context_line":""},{"line_number":246,"context_line":"Cinder-client impact"},{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"7a77a97e_e9338811","line":246,"range":{"start_line":246,"start_character":0,"end_line":246,"end_character":20},"updated":"2016-11-22 18:21:21.000000000","message":"Please also include OSC implementation as we are moving towards full compatibility between both:\nopenstack snapshot revert --force \u003csnapshot\u003e","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2e030971ad8e247060050680eb2dedc8f8f7fd9d","unresolved":false,"context_lines":[{"line_number":243,"context_line":"Notifications about the revert process(start, end, error) and"},{"line_number":244,"context_line":"so does the related change notification will be add."},{"line_number":245,"context_line":""},{"line_number":246,"context_line":"Cinder-client impact"},{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_fea47009","line":246,"range":{"start_line":246,"start_character":0,"end_line":246,"end_character":20},"in_reply_to":"7a77a97e_e9338811","updated":"2016-11-23 03:25:05.000000000","message":"Done","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"882abe724abee49c1f0e827afcde4a4e538cc237","unresolved":false,"context_lines":[{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""},{"line_number":250,"context_line":"  cinder revert-to-snapshot \u003csnapshot\u003e \u003c--force\u003e"},{"line_number":251,"context_line":""},{"line_number":252,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":253,"context_line":"  \u003csnapshot\u003e  Name or ID of the snapshot to restore."}],"source_content_type":"text/x-rst","patch_set":19,"id":"7a77a97e_29d4a0bb","line":250,"range":{"start_line":250,"start_character":2,"end_line":250,"end_character":48},"updated":"2016-11-22 18:21:21.000000000","message":"This should follow the Cinder standard (there\u0027s actually no standard but lets keep consistency):\ncinder snapshot-revert \u003csnapshot\u003e [--force]","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":11865,"name":"Clinton Knight","email":"cknight@netapp.com","username":"cknight"},"change_message_id":"303bfa0cbc36e07550416ad4afe37930a56a985d","unresolved":false,"context_lines":[{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""},{"line_number":250,"context_line":"  cinder revert-to-snapshot \u003csnapshot\u003e \u003c--force\u003e"},{"line_number":251,"context_line":""},{"line_number":252,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":253,"context_line":"  \u003csnapshot\u003e  Name or ID of the snapshot to restore."}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_2997f948","line":250,"range":{"start_line":250,"start_character":2,"end_line":250,"end_character":48},"in_reply_to":"5a74a57a_143e09b8","updated":"2016-11-27 15:58:22.000000000","message":"The *volume* is being reverted, not the snapshot, so the full CLI name would be \"volume-revert-to-snapshot\".  But the precedent is to omit the \"volume\" prefix for simple commands (\"list\", \"show\", \"create\", \"manage\", \"delete\", etc.), hence the conscious choice of \"revert-to-snapshot\".","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"1e2e925bf8fd2b7071b63e20d60b8742a7e507b1","unresolved":false,"context_lines":[{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""},{"line_number":250,"context_line":"  cinder revert-to-snapshot \u003csnapshot\u003e \u003c--force\u003e"},{"line_number":251,"context_line":""},{"line_number":252,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":253,"context_line":"  \u003csnapshot\u003e  Name or ID of the snapshot to restore."}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_97ae5ae4","line":250,"range":{"start_line":250,"start_character":2,"end_line":250,"end_character":48},"in_reply_to":"5a74a57a_2997f948","updated":"2016-11-28 09:46:36.000000000","message":"Ok it makes sense!","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"63924a0a98e96fd3b68e8986e235935fcdcb5cf5","unresolved":false,"context_lines":[{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""},{"line_number":250,"context_line":"  cinder revert-to-snapshot \u003csnapshot\u003e \u003c--force\u003e"},{"line_number":251,"context_line":""},{"line_number":252,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":253,"context_line":"  \u003csnapshot\u003e  Name or ID of the snapshot to restore."}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_143e09b8","line":250,"range":{"start_line":250,"start_character":2,"end_line":250,"end_character":48},"in_reply_to":"5a74a57a_fe1f1059","updated":"2016-11-25 04:23:05.000000000","message":"Hmmm, I missed this in the review I did there. Ill see with then if they can use snapshot revert.","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2e030971ad8e247060050680eb2dedc8f8f7fd9d","unresolved":false,"context_lines":[{"line_number":247,"context_line":"--------------------"},{"line_number":248,"context_line":"The cinder-client will add commands to expose the revert API::"},{"line_number":249,"context_line":""},{"line_number":250,"context_line":"  cinder revert-to-snapshot \u003csnapshot\u003e \u003c--force\u003e"},{"line_number":251,"context_line":""},{"line_number":252,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":253,"context_line":"  \u003csnapshot\u003e  Name or ID of the snapshot to restore."}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_fe1f1059","line":250,"range":{"start_line":250,"start_character":2,"end_line":250,"end_character":48},"in_reply_to":"7a77a97e_29d4a0bb","updated":"2016-11-23 03:25:05.000000000","message":"Manila use  \u0027revert-to-snapshot\u0027 instead of \u0027snapshot-revert\u0027 [1], which one should we take? suggest we leave \u0027revert-to-snapshot\u0027 here and use snapshot revert in OSC\n\n[1]  https://review.openstack.org/#/c/391269/6/specs/ocata/manila-share-revert-to-snapshot.rst","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"882abe724abee49c1f0e827afcde4a4e538cc237","unresolved":false,"context_lines":[{"line_number":254,"context_line":"  \u003c--force\u003e   whether revert an attached volume."},{"line_number":255,"context_line":"Other end user impact"},{"line_number":256,"context_line":"---------------------"},{"line_number":257,"context_line":"This feature would also be implemented to cinder-ui, but should restricted"},{"line_number":258,"context_line":"to the latest snapshot resource."},{"line_number":259,"context_line":""},{"line_number":260,"context_line":"None"},{"line_number":261,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"7a77a97e_298660db","line":258,"range":{"start_line":257,"start_character":53,"end_line":258,"end_character":32},"updated":"2016-11-22 18:21:21.000000000","message":"This is restricted to the latest snapshot in any situation, so no need to mention this here.","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2e030971ad8e247060050680eb2dedc8f8f7fd9d","unresolved":false,"context_lines":[{"line_number":254,"context_line":"  \u003c--force\u003e   whether revert an attached volume."},{"line_number":255,"context_line":"Other end user impact"},{"line_number":256,"context_line":"---------------------"},{"line_number":257,"context_line":"This feature would also be implemented to cinder-ui, but should restricted"},{"line_number":258,"context_line":"to the latest snapshot resource."},{"line_number":259,"context_line":""},{"line_number":260,"context_line":"None"},{"line_number":261,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_8a55f71e","line":258,"range":{"start_line":257,"start_character":53,"end_line":258,"end_character":32},"in_reply_to":"7a77a97e_298660db","updated":"2016-11-23 03:25:05.000000000","message":"Done","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"882abe724abee49c1f0e827afcde4a4e538cc237","unresolved":false,"context_lines":[{"line_number":305,"context_line":"- Implement Cinder UI"},{"line_number":306,"context_line":"- Add API documents"},{"line_number":307,"context_line":"- Update develop reference and support matrix"},{"line_number":308,"context_line":""},{"line_number":309,"context_line":"Dependencies"},{"line_number":310,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":311,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"7a77a97e_0c6e12c9","line":308,"updated":"2016-11-22 18:21:21.000000000","message":"- Implement OSC","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"2e030971ad8e247060050680eb2dedc8f8f7fd9d","unresolved":false,"context_lines":[{"line_number":305,"context_line":"- Implement Cinder UI"},{"line_number":306,"context_line":"- Add API documents"},{"line_number":307,"context_line":"- Update develop reference and support matrix"},{"line_number":308,"context_line":""},{"line_number":309,"context_line":"Dependencies"},{"line_number":310,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":311,"context_line":""}],"source_content_type":"text/x-rst","patch_set":19,"id":"5a74a57a_2a25c3c1","line":308,"in_reply_to":"7a77a97e_0c6e12c9","updated":"2016-11-23 03:25:05.000000000","message":"done","commit_id":"7c990ce0b0d5f0666b00591d8fe98661e9d1ab3a"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"de291891145813c53360b36a700b87ba263b8e64","unresolved":false,"context_lines":[{"line_number":46,"context_line":"7. User attach new volumes"},{"line_number":47,"context_line":""},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"* By creating volume from snapshot, some of vendors need to copy data to the"},{"line_number":50,"context_line":"  new volume, It takes more time to copy the data. But the end user hopes that"},{"line_number":51,"context_line":"  the instance can recover quickly. Also, the new volume could not keep the"},{"line_number":52,"context_line":"  same UUID(so does the disk UUID, host...), in the system, but several"},{"line_number":53,"context_line":"  applications use that to boot. By reverting volume from snapshot, it will"},{"line_number":54,"context_line":"  keep the same volume id  and makes reverting a instance much easier since"},{"line_number":55,"context_line":"  device names inside the instance stay the same."},{"line_number":56,"context_line":""},{"line_number":57,"context_line":"* By Restoring from backup, restoring backup data to volume may be a good way"},{"line_number":58,"context_line":"  to recover data when the storage does not exist, but it\u0027s not a common case"}],"source_content_type":"text/x-rst","patch_set":20,"id":"1a6eadb0_68192688","line":55,"range":{"start_line":49,"start_character":2,"end_line":55,"end_character":49},"updated":"2016-12-14 12:20:32.000000000","message":"I believe I have mentioned about the possibility of this being a problem in a mailing list in the past. I have tested this and verified that this didn\u0027t happen in my environment (the UUID is stored inside de block device and is kept). Do you have other references about this?","commit_id":"ae9133526aedfe460352d8d8f14a05c8cf8740b7"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"25539dbc54b9452f4d2147623ef0294cfdb594db","unresolved":false,"context_lines":[{"line_number":192,"context_line":"If the cinder driver can not support the revert snapshot feature,"},{"line_number":193,"context_line":"the framework will use the general way to implement the revert snapshot"},{"line_number":194,"context_line":"feature."},{"line_number":195,"context_line":""},{"line_number":196,"context_line":"1. create a hidden volume from snapshot"},{"line_number":197,"context_line":"2. mount hidden volume to host"},{"line_number":198,"context_line":"3. copy data from hidden volume to original volume"}],"source_content_type":"text/x-rst","patch_set":20,"id":"5a74a57a_6fec682a","line":195,"updated":"2016-11-28 17:21:52.000000000","message":"I think this does not work, for the reasons noted on patchset 18.","commit_id":"ae9133526aedfe460352d8d8f14a05c8cf8740b7"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"de291891145813c53360b36a700b87ba263b8e64","unresolved":false,"context_lines":[{"line_number":192,"context_line":"If the cinder driver can not support the revert snapshot feature,"},{"line_number":193,"context_line":"the framework will use the general way to implement the revert snapshot"},{"line_number":194,"context_line":"feature."},{"line_number":195,"context_line":""},{"line_number":196,"context_line":"1. create a hidden volume from snapshot"},{"line_number":197,"context_line":"2. mount hidden volume to host"},{"line_number":198,"context_line":"3. copy data from hidden volume to original volume"}],"source_content_type":"text/x-rst","patch_set":20,"id":"1a6eadb0_8b93dc4e","line":195,"in_reply_to":"5a74a57a_6fec682a","updated":"2016-12-14 12:20:32.000000000","message":"@Eric, your concern is not clear to me. This temporary volume that will be created is only to be able to get the data from the snapshot. This volume does not need to keep all snapshot tree of the volume we want to revert (that one will keep all its snapshots). Am I missing something?","commit_id":"ae9133526aedfe460352d8d8f14a05c8cf8740b7"},{"author":{"_account_id":18967,"name":"zhangsong","email":"zhangsong@cmss.chinamobile.com","username":"zhangsong"},"change_message_id":"73ce8ada65f28c19c6d10f52022aa0593f3b7070","unresolved":false,"context_lines":[{"line_number":228,"context_line":"is intended for both tenants and admins to use, and the policy.json"},{"line_number":229,"context_line":"file will be updated to reflect allowed use by all.  If the snapshot is"},{"line_number":230,"context_line":"not the most recent one known to cinder, or the state of either volume or"},{"line_number":231,"context_line":"snapshot is not \u0027available\u0027, the API will return the HTTP error code 409"},{"line_number":232,"context_line":"(Conflict).  If the volume isn\u0027t found, 404.  If the snapshot doesn\u0027t exist,"},{"line_number":233,"context_line":"400 (because it isn\u0027t explicitly referenced in the URL)."},{"line_number":234,"context_line":""}],"source_content_type":"text/x-rst","patch_set":20,"id":"3a71b18c_88000b04","line":231,"range":{"start_line":231,"start_character":0,"end_line":231,"end_character":8},"updated":"2016-12-12 02:43:16.000000000","message":"Hi, I have some suggestion:\n1.I don\u0027t think we need \u0027snapshot_id\u0027 if the volume can only be reverted to the most recent snapshot; \n2.I think the most convenient way is to revert volume to the specified snapshot, and use the most recent one as default if user don\u0027t specify it;","commit_id":"ae9133526aedfe460352d8d8f14a05c8cf8740b7"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"In order to support revert an attached volume, user should set an \u0027force\u0027"},{"line_number":122,"context_line":"flag with which cinder can shutdown the instance(by nova API) automatically"},{"line_number":123,"context_line":"and revert volume in \u0027attached\u0027 status."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Therefore, we need to do the following changes in order to support volume"},{"line_number":126,"context_line":"revert to snapshot."}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_0046bdbe","line":123,"updated":"2017-01-04 15:45:18.000000000","message":"\"force\" is not a good name for a field here.\n\nBut, I don\u0027t know that shutting down the instance on a revert request is reasonable anyway.  I think it would be better (at least for the first implementation) to just reject the request if it\u0027s attached.\n\nWhat happens if the volume is being backed up rather than is attached to an instance?","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"ef1f8ce14c738ab51d0d19f74537fdf920b15951","unresolved":false,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"In order to support revert an attached volume, user should set an \u0027force\u0027"},{"line_number":122,"context_line":"flag with which cinder can shutdown the instance(by nova API) automatically"},{"line_number":123,"context_line":"and revert volume in \u0027attached\u0027 status."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Therefore, we need to do the following changes in order to support volume"},{"line_number":126,"context_line":"revert to snapshot."}],"source_content_type":"text/x-rst","patch_set":21,"id":"da36d5c6_8dc39856","line":123,"in_reply_to":"ba5201f7_0046bdbe","updated":"2017-02-23 19:13:15.000000000","message":"is there any change in the status of the volume when is being backed up? What happens if the user deletes the volume in the middle of a backup?","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":129,"context_line":""},{"line_number":130,"context_line":"    URL \u003d /v3/{tenant_id}/volumes/{volume_id}/action"},{"line_number":131,"context_line":"    Method: POST"},{"line_number":132,"context_line":"    Body \u003d {\u0027revert\u0027: {\u0027snapshot_id\u0027: \u003csnapshot_id\u003e, \u0027force\u0027: \u003cTrue|False\u003e}}"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"    In the API layer, the validations(existence and status) on both"},{"line_number":135,"context_line":"    snapshot and volume will be proceeded along with whether the snapshot is"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_8010eda9","line":132,"updated":"2017-01-04 15:45:18.000000000","message":"Remove \"force\".","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":154,"context_line":"    After a successful restoration, both objects\u0027 status will again be set"},{"line_number":155,"context_line":"    to the status before reverting.  If the restoration fails, the volume"},{"line_number":156,"context_line":"    (both attached and detached) will be set to \u0027reverting_error\u0027, and the"},{"line_number":157,"context_line":"    snapshot will be set to \u0027available\u0027."},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"* Add a function that vendors can use to override the default"},{"line_number":160,"context_line":"revert_to_snapshot function with an optimized version. Vendors can implement"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_e04ff9ba","line":157,"updated":"2017-01-04 15:45:18.000000000","message":"I don\u0027t think we should introduce a new \"reverting_error\" state.  It would be better to use \"error\" and fire an async error message indicating what failure occurred.","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":20382,"name":"YuYang","email":"yuyangwang1985@gmail.com","username":"YuYangWang"},"change_message_id":"8e80068bcf9ff8520f672002f1ab6cf7de921b2d","unresolved":false,"context_lines":[{"line_number":177,"context_line":"the framework will use the generic way to implement the revert snapshot"},{"line_number":178,"context_line":"feature."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"1. create a hidden volume from snapshot"},{"line_number":181,"context_line":"2. mount hidden volume to host"},{"line_number":182,"context_line":"3. copy data from hidden volume to original volume"},{"line_number":183,"context_line":"4. delete the hidden volume"}],"source_content_type":"text/x-rst","patch_set":21,"id":"da4df55a_cf37b9ae","line":180,"range":{"start_line":180,"start_character":12,"end_line":180,"end_character":25},"updated":"2016-12-23 03:34:00.000000000","message":"how to create hidden volume?thanks!","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":177,"context_line":"the framework will use the generic way to implement the revert snapshot"},{"line_number":178,"context_line":"feature."},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"1. create a hidden volume from snapshot"},{"line_number":181,"context_line":"2. mount hidden volume to host"},{"line_number":182,"context_line":"3. copy data from hidden volume to original volume"},{"line_number":183,"context_line":"4. delete the hidden volume"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_80d54dba","line":180,"range":{"start_line":180,"start_character":12,"end_line":180,"end_character":25},"in_reply_to":"da4df55a_cf37b9ae","updated":"2017-01-04 15:45:18.000000000","message":"What does \"hidden\" mean here?  Owned by the Cinder internal tenant?","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":179,"context_line":""},{"line_number":180,"context_line":"1. create a hidden volume from snapshot"},{"line_number":181,"context_line":"2. mount hidden volume to host"},{"line_number":182,"context_line":"3. copy data from hidden volume to original volume"},{"line_number":183,"context_line":"4. delete the hidden volume"},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"Alternatives"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_20ef4188","line":182,"updated":"2017-01-04 15:45:18.000000000","message":"What if the original volume is smaller than the current snapshot because the volume was extended?","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":195,"context_line":"-----------------"},{"line_number":196,"context_line":""},{"line_number":197,"context_line":"New status \u0027reverting\u0027, \u0027reverting_error\u0027 to volume object and new status"},{"line_number":198,"context_line":"\u0027restoring\u0027 to snapshot object."},{"line_number":199,"context_line":""},{"line_number":200,"context_line":"REST API impact"},{"line_number":201,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_80ec6d90","line":198,"updated":"2017-01-04 15:45:18.000000000","message":"I don\u0027t think we should introduce \"reverting_error\".","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":20382,"name":"YuYang","email":"yuyangwang1985@gmail.com","username":"YuYangWang"},"change_message_id":"8e80068bcf9ff8520f672002f1ab6cf7de921b2d","unresolved":false,"context_lines":[{"line_number":205,"context_line":""},{"line_number":206,"context_line":"  URL \u003d /v3/{tenant_id}/volumes/{volume_id}/action"},{"line_number":207,"context_line":"  Method: POST"},{"line_number":208,"context_line":"  Body \u003d {\u0027revert\u0027: {\u0027snapshot_id\u0027: \u003csnapshot_id\u003e, \u0027force\u0027: \u003cTrue|False\u003e}}"},{"line_number":209,"context_line":"  Normal Response: Status code 202 with empty response body"},{"line_number":210,"context_line":""},{"line_number":211,"context_line":"Calling this method reverts a volume to the specified snapshot.  It"}],"source_content_type":"text/x-rst","patch_set":21,"id":"da4df55a_6f6ecd97","line":208,"range":{"start_line":208,"start_character":61,"end_line":208,"end_character":71},"updated":"2016-12-23 03:34:00.000000000","message":"if add at what time set True or False？an default?thanks!","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":235,"context_line":""},{"line_number":236,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":237,"context_line":"  snapshot  Name or ID of the snapshot to restore."},{"line_number":238,"context_line":"  --force   whether revert an attached volume."},{"line_number":239,"context_line":""},{"line_number":240,"context_line":"OpenStack-client impact"},{"line_number":241,"context_line":"-----------------------"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_00f95d4d","line":238,"updated":"2017-01-04 15:45:18.000000000","message":"Should not use \"--force\" for this.","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7ec9a2f5018766017a749ac54ff86dbdac0ea1c","unresolved":false,"context_lines":[{"line_number":246,"context_line":""},{"line_number":247,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":248,"context_line":"  snapshot  Name or ID of the snapshot to restore."},{"line_number":249,"context_line":"  --force   whether revert an attached volume."},{"line_number":250,"context_line":""},{"line_number":251,"context_line":"Other end user impact"},{"line_number":252,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":21,"id":"ba5201f7_60a06944","line":249,"updated":"2017-01-04 15:45:18.000000000","message":"Should not use \"--force\" for this.","commit_id":"99a2f9f606b180f51005c7e882c4eeea77868c65"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"3559868c9eb9d70eb56dedeb067b8db0e636489e","unresolved":false,"context_lines":[{"line_number":259,"context_line":"    Body \u003d {"},{"line_number":260,"context_line":"               \u0027revert\u0027: {"},{"line_number":261,"context_line":"                   \u0027snapshot_id\u0027: \u003csnapshot_id\u003e,"},{"line_number":262,"context_line":"                   \u0027attached\u0027: \u003cTrue|False\u003e"},{"line_number":263,"context_line":"               }"},{"line_number":264,"context_line":"           }"},{"line_number":265,"context_line":""}],"source_content_type":"text/x-rst","patch_set":23,"id":"da36d5c6_299c5b0e","line":262,"updated":"2017-02-23 20:46:41.000000000","message":"\"attached\" should not be a field here -- this should only support reverts for volumes that are detached.","commit_id":"7a172823ab660ac2ee071e6653ed1dd67780489a"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"91b6c938e08237962737dd3c7e84c69d3b60b428","unresolved":false,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"There are two cases when reverting a volume, one is detached volume or which"},{"line_number":122,"context_line":"can be detached, and the other is the bootable volume which can\u0027t be detached."},{"line_number":123,"context_line":"We would only support reverting a detached volume at present. But also leave"},{"line_number":124,"context_line":"the API easy to be extended for bootable volumes. So if user would like to"},{"line_number":125,"context_line":"revert a volume. He should the detach the volume first."},{"line_number":126,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_c754df55","line":123,"range":{"start_line":123,"start_character":0,"end_line":123,"end_character":60},"updated":"2017-04-27 14:01:18.000000000","message":"+1","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"8e8fccfd0bed383123111f4e727cf6c0f4eac9a0","unresolved":false,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"There are two cases when reverting a volume, one is detached volume or which"},{"line_number":122,"context_line":"can be detached, and the other is the bootable volume which can\u0027t be detached."},{"line_number":123,"context_line":"We would only support reverting a detached volume at present. But also leave"},{"line_number":124,"context_line":"the API easy to be extended for bootable volumes. So if user would like to"},{"line_number":125,"context_line":"revert a volume. He should the detach the volume first."},{"line_number":126,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_d1b5b833","line":123,"range":{"start_line":123,"start_character":50,"end_line":123,"end_character":60},"updated":"2017-04-27 16:20:33.000000000","message":"I\u0027m not sure what \"at present\" implies here -- I don\u0027t see how you can ever revert a volume that\u0027s attached.\n\nAnything using that volume would just break.","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"a3e469d42ba0405ac4cafeefe3c7f576ab9631e9","unresolved":false,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"There are two cases when reverting a volume, one is detached volume or which"},{"line_number":122,"context_line":"can be detached, and the other is the bootable volume which can\u0027t be detached."},{"line_number":123,"context_line":"We would only support reverting a detached volume at present. But also leave"},{"line_number":124,"context_line":"the API easy to be extended for bootable volumes. So if user would like to"},{"line_number":125,"context_line":"revert a volume. He should the detach the volume first."},{"line_number":126,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_6bc33bee","line":123,"range":{"start_line":123,"start_character":50,"end_line":123,"end_character":60},"in_reply_to":"5ff73747_d1b5b833","updated":"2017-05-02 02:53:12.000000000","message":"You are right,Eric, we could only support revert a volume that is detached.","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"8e8fccfd0bed383123111f4e727cf6c0f4eac9a0","unresolved":false,"context_lines":[{"line_number":121,"context_line":"There are two cases when reverting a volume, one is detached volume or which"},{"line_number":122,"context_line":"can be detached, and the other is the bootable volume which can\u0027t be detached."},{"line_number":123,"context_line":"We would only support reverting a detached volume at present. But also leave"},{"line_number":124,"context_line":"the API easy to be extended for bootable volumes. So if user would like to"},{"line_number":125,"context_line":"revert a volume. He should the detach the volume first."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"As we support extend volumes at present, we could meet the case that"}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_b1b22c48","line":124,"range":{"start_line":124,"start_character":32,"end_line":124,"end_character":48},"updated":"2017-04-27 16:20:33.000000000","message":"Why would the current proposal not handle bootable volumes?","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"a3e469d42ba0405ac4cafeefe3c7f576ab9631e9","unresolved":false,"context_lines":[{"line_number":121,"context_line":"There are two cases when reverting a volume, one is detached volume or which"},{"line_number":122,"context_line":"can be detached, and the other is the bootable volume which can\u0027t be detached."},{"line_number":123,"context_line":"We would only support reverting a detached volume at present. But also leave"},{"line_number":124,"context_line":"the API easy to be extended for bootable volumes. So if user would like to"},{"line_number":125,"context_line":"revert a volume. He should the detach the volume first."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"As we support extend volumes at present, we could meet the case that"}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_4bb63f8c","line":124,"range":{"start_line":124,"start_character":32,"end_line":124,"end_character":48},"in_reply_to":"5ff73747_b1b22c48","updated":"2017-05-02 02:53:12.000000000","message":"This should be attached volumes.","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"91b6c938e08237962737dd3c7e84c69d3b60b428","unresolved":false,"context_lines":[{"line_number":155,"context_line":"  use the generic implementation to run revert process."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"  After a successful restoration, both objects\u0027 status will again be set"},{"line_number":158,"context_line":"  to the status before reverting.  If the restoration fails, the volume"},{"line_number":159,"context_line":"  (both attached and detached) will be set to \u0027error\u0027, and the"},{"line_number":160,"context_line":"  snapshot will be set to \u0027available\u0027."},{"line_number":161,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_276b83f1","line":158,"updated":"2017-04-27 14:01:18.000000000","message":"Since we are now only allowing revert of \u0027available\u0027 volumes, I think some of this can be removed.\n\nI have a concern about the \u0027error\u0027 status though. If a user requests a volume to be reverted and it fails, how do they get out of this error state?","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"a3e469d42ba0405ac4cafeefe3c7f576ab9631e9","unresolved":false,"context_lines":[{"line_number":155,"context_line":"  use the generic implementation to run revert process."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"  After a successful restoration, both objects\u0027 status will again be set"},{"line_number":158,"context_line":"  to the status before reverting.  If the restoration fails, the volume"},{"line_number":159,"context_line":"  (both attached and detached) will be set to \u0027error\u0027, and the"},{"line_number":160,"context_line":"  snapshot will be set to \u0027available\u0027."},{"line_number":161,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_06e94ee4","line":158,"in_reply_to":"5ff73747_276b83f1","updated":"2017-05-02 02:53:12.000000000","message":"Thanks, I will simplify this section.","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"91b6c938e08237962737dd3c7e84c69d3b60b428","unresolved":false,"context_lines":[{"line_number":175,"context_line":"          :return None"},{"line_number":176,"context_line":"          \"\"\""},{"line_number":177,"context_line":""},{"line_number":178,"context_line":"  Exception will be catched and stored if any inner error raised from"},{"line_number":179,"context_line":"  this function."},{"line_number":180,"context_line":""},{"line_number":181,"context_line":"- **Add a generic revert to snapshot implementation function**."}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_8771370f","line":178,"range":{"start_line":178,"start_character":20,"end_line":178,"end_character":27},"updated":"2017-04-27 14:01:18.000000000","message":"caught","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"a3e469d42ba0405ac4cafeefe3c7f576ab9631e9","unresolved":false,"context_lines":[{"line_number":175,"context_line":"          :return None"},{"line_number":176,"context_line":"          \"\"\""},{"line_number":177,"context_line":""},{"line_number":178,"context_line":"  Exception will be catched and stored if any inner error raised from"},{"line_number":179,"context_line":"  this function."},{"line_number":180,"context_line":""},{"line_number":181,"context_line":"- **Add a generic revert to snapshot implementation function**."}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_aba4a34b","line":178,"range":{"start_line":178,"start_character":20,"end_line":178,"end_character":27},"in_reply_to":"5ff73747_8771370f","updated":"2017-05-02 02:53:12.000000000","message":"Done","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"8e8fccfd0bed383123111f4e727cf6c0f4eac9a0","unresolved":false,"context_lines":[{"line_number":251,"context_line":"    1. 404, if the volume is not found."},{"line_number":252,"context_line":"    2. 409, if volume and snapshot\u0027s status are not \u0027available\u0027or"},{"line_number":253,"context_line":"       the sizes of volume and snapshot are not equal."},{"line_number":254,"context_line":"    3. 400, if the snapshot is not found or not belongs to the specified"},{"line_number":255,"context_line":"       volume or is not the latest one."},{"line_number":256,"context_line":""},{"line_number":257,"context_line":"Calling this method reverts a volume to the specified snapshot."}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_31861c15","line":254,"range":{"start_line":254,"start_character":19,"end_line":254,"end_character":40},"updated":"2017-04-27 16:20:33.000000000","message":"Snapshot not found should probably also be 404?","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"a3e469d42ba0405ac4cafeefe3c7f576ab9631e9","unresolved":false,"context_lines":[{"line_number":251,"context_line":"    1. 404, if the volume is not found."},{"line_number":252,"context_line":"    2. 409, if volume and snapshot\u0027s status are not \u0027available\u0027or"},{"line_number":253,"context_line":"       the sizes of volume and snapshot are not equal."},{"line_number":254,"context_line":"    3. 400, if the snapshot is not found or not belongs to the specified"},{"line_number":255,"context_line":"       volume or is not the latest one."},{"line_number":256,"context_line":""},{"line_number":257,"context_line":"Calling this method reverts a volume to the specified snapshot."}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_ab7943c5","line":254,"range":{"start_line":254,"start_character":19,"end_line":254,"end_character":40},"in_reply_to":"5ff73747_31861c15","updated":"2017-05-02 02:53:12.000000000","message":"I reckon 400 is suitable here, the snapshot resource id is not specified in our request url.","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"8e8fccfd0bed383123111f4e727cf6c0f4eac9a0","unresolved":false,"context_lines":[{"line_number":284,"context_line":""},{"line_number":285,"context_line":"The openstack-client will add command to expose the revert API::"},{"line_number":286,"context_line":""},{"line_number":287,"context_line":"  usage: openstack volume snapshot restore \u003csnapshot\u003e \u003cattached\u003e"},{"line_number":288,"context_line":""},{"line_number":289,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":290,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_713dc48d","line":287,"range":{"start_line":287,"start_character":54,"end_line":287,"end_character":64},"updated":"2017-04-27 16:20:33.000000000","message":"What is the attached arg?","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":23083,"name":"TommyLike","email":"tommylikehu@gmail.com","username":"TommyLike"},"change_message_id":"a3e469d42ba0405ac4cafeefe3c7f576ab9631e9","unresolved":false,"context_lines":[{"line_number":284,"context_line":""},{"line_number":285,"context_line":"The openstack-client will add command to expose the revert API::"},{"line_number":286,"context_line":""},{"line_number":287,"context_line":"  usage: openstack volume snapshot restore \u003csnapshot\u003e \u003cattached\u003e"},{"line_number":288,"context_line":""},{"line_number":289,"context_line":"  Revert a volume to the specified snapshot."},{"line_number":290,"context_line":""}],"source_content_type":"text/x-rst","patch_set":24,"id":"5ff73747_eb5ceb4b","line":287,"range":{"start_line":287,"start_character":54,"end_line":287,"end_character":64},"in_reply_to":"5ff73747_713dc48d","updated":"2017-05-02 02:53:12.000000000","message":"Thanks","commit_id":"3da8e6dc3bb38666720bce3d03ec9a2b0e8a1e22"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"bfc8f4e15ae3bc0f9b1b9756b7e58e8fe0804bfe","unresolved":false,"context_lines":[{"line_number":134,"context_line":""},{"line_number":135,"context_line":"- **Add volume revert to snapshot API**."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"  We will add a revert to snapshot API, In the API layer, the validations ("},{"line_number":138,"context_line":"  existence, status and size) on both snapshot and volume will be proceeded"},{"line_number":139,"context_line":"  along with whether the snapshot is the latest one."},{"line_number":140,"context_line":""}],"source_content_type":"text/x-rst","patch_set":25,"id":"5ff73747_8c744f9e","line":137,"range":{"start_line":137,"start_character":40,"end_line":137,"end_character":42},"updated":"2017-05-02 13:35:22.000000000","message":"nit: Only if you need to make an update: s/In/in/","commit_id":"7323f8c7ca6c41665410696a24004e3e97b92a6f"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"bfc8f4e15ae3bc0f9b1b9756b7e58e8fe0804bfe","unresolved":false,"context_lines":[{"line_number":230,"context_line":"    1. 404, if the volume is not found."},{"line_number":231,"context_line":"    2. 409, if volume and snapshot\u0027s status are not \u0027available\u0027or"},{"line_number":232,"context_line":"       the sizes of volume and snapshot are not equal."},{"line_number":233,"context_line":"    3. 400, if the snapshot is not found or not belongs to the specified"},{"line_number":234,"context_line":"       volume or is not the latest one."},{"line_number":235,"context_line":""},{"line_number":236,"context_line":"Calling this method reverts a volume to the specified snapshot."}],"source_content_type":"text/x-rst","patch_set":25,"id":"5ff73747_4cf1a701","line":233,"updated":"2017-05-02 13:35:22.000000000","message":"I think BAD REQUEST is OK here.","commit_id":"7323f8c7ca6c41665410696a24004e3e97b92a6f"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"33e49213333c20c59dc1117da55630a1c4b1baee","unresolved":false,"context_lines":[{"line_number":56,"context_line":"* It consumes quota from the user and in some cases real space on the backend."},{"line_number":57,"context_line":""},{"line_number":58,"context_line":"Another alternative is to restore the data from a backup, but that process"},{"line_number":59,"context_line":"can\u0027t be done by the cloud user, it it\u0027s not a common case (data inconsistency"},{"line_number":60,"context_line":"is more likely the main cause), also, the end user would expect a less time"},{"line_number":61,"context_line":"consuming operation."},{"line_number":62,"context_line":""}],"source_content_type":"text/x-rst","patch_set":26,"id":"5ff73747_b7451846","line":59,"range":{"start_line":59,"start_character":33,"end_line":59,"end_character":38},"updated":"2017-05-02 14:00:11.000000000","message":"nit: duplicate \"it\".","commit_id":"45a716ebc034ce803e7f3f1c4d8b7e9a5fdd6f4e"},{"author":{"_account_id":6491,"name":"xing-yang","email":"xingyang105@gmail.com","username":"xing-yang"},"change_message_id":"33e49213333c20c59dc1117da55630a1c4b1baee","unresolved":false,"context_lines":[{"line_number":187,"context_line":"  the framework will use the generic way to implement the revert snapshot"},{"line_number":188,"context_line":"  feature."},{"line_number":189,"context_line":""},{"line_number":190,"context_line":"  1. create a temporary volume from snapshot."},{"line_number":191,"context_line":"  2. mount temporary volume to host."},{"line_number":192,"context_line":"  3. copy data from temporary volume to original volume."},{"line_number":193,"context_line":"  4. delete the temporary volume."},{"line_number":194,"context_line":""},{"line_number":195,"context_line":"Alternatives"},{"line_number":196,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":26,"id":"5ff73747_173f2cb6","line":193,"range":{"start_line":190,"start_character":5,"end_line":193,"end_character":33},"updated":"2017-05-02 14:00:11.000000000","message":"If driver supports attach snapshot, should directly use snapshot like in the backup snapshot case.","commit_id":"45a716ebc034ce803e7f3f1c4d8b7e9a5fdd6f4e"}]}
