)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"ec4aa31880b73aa07a087a1af4d46c0f8fab8c93","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Driver reinitialization after failure"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This is a proposal to reinitialize a volume driver"},{"line_number":10,"context_line":"when it fails during starting."},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"Change-Id: Ic8b58efb240efa535cc5cb039139b497b48f1570"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"3f79a3b5_f8258216","line":9,"range":{"start_line":9,"start_character":0,"end_line":9,"end_character":4},"updated":"2018-12-13 11:54:33.000000000","message":"It would be nice to point out here that this is an update to the spec after the discussion that we had on the Cinder meeting.","commit_id":"f28dd336c69f3facbee8d4daf8f568276ce707ee"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"a1fae2c34c0c8b7473cea90d96e66215b7628006","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Driver reinitialization after failure"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This is a proposal to reinitialize a volume driver"},{"line_number":10,"context_line":"when it fails during starting."},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"Change-Id: Ic8b58efb240efa535cc5cb039139b497b48f1570"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"3f79a3b5_ce09e0db","line":9,"range":{"start_line":9,"start_character":0,"end_line":9,"end_character":4},"in_reply_to":"3f79a3b5_f8258216","updated":"2018-12-13 12:47:10.000000000","message":"Done","commit_id":"f28dd336c69f3facbee8d4daf8f568276ce707ee"}],"specs/stein/driver-reinitialization-after-fail.rst":[{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"ec4aa31880b73aa07a087a1af4d46c0f8fab8c93","unresolved":false,"context_lines":[{"line_number":16,"context_line":"Problem description"},{"line_number":17,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":18,"context_line":""},{"line_number":19,"context_line":"When users use some tool to automatically deploy their Cinder environment,"},{"line_number":20,"context_line":"sometimes when Cinder volume service starts, its backend storage is not"},{"line_number":21,"context_line":"ready. As a result, Cinder volume service can\u0027t initialize the volume"},{"line_number":22,"context_line":"driver, but the volume service process still exists, running with periodly"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_b835caf7","line":19,"range":{"start_line":19,"start_character":0,"end_line":19,"end_character":4},"updated":"2018-12-13 11:54:33.000000000","message":"A more precise statement of the problem is: During Cinder initialization, for many reasons, the storage backend might not be ready and responding. In this case, the driver will not be loaded even if the array gets available right after. As there\u0027s no...","commit_id":"f28dd336c69f3facbee8d4daf8f568276ce707ee"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"a1fae2c34c0c8b7473cea90d96e66215b7628006","unresolved":false,"context_lines":[{"line_number":16,"context_line":"Problem description"},{"line_number":17,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":18,"context_line":""},{"line_number":19,"context_line":"When users use some tool to automatically deploy their Cinder environment,"},{"line_number":20,"context_line":"sometimes when Cinder volume service starts, its backend storage is not"},{"line_number":21,"context_line":"ready. As a result, Cinder volume service can\u0027t initialize the volume"},{"line_number":22,"context_line":"driver, but the volume service process still exists, running with periodly"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_ae04a4c5","line":19,"range":{"start_line":19,"start_character":0,"end_line":19,"end_character":4},"in_reply_to":"3f79a3b5_b835caf7","updated":"2018-12-13 12:47:10.000000000","message":"Done","commit_id":"f28dd336c69f3facbee8d4daf8f568276ce707ee"},{"author":{"_account_id":10058,"name":"Erlon R. Cruz","email":"erlon.rodrigues.cruz@canonical.com","username":"sombrafam"},"change_message_id":"ec4aa31880b73aa07a087a1af4d46c0f8fab8c93","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Allow reinitialization of a volume driver when it failed to initialize."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- Import a configuration to set the maximum retry numbers."},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"- About the inveral of retry. The interval of publishing volume driver\u0027s"},{"line_number":47,"context_line":"  capability will be used as the original interval. Later every time if"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_530fb72b","line":44,"range":{"start_line":44,"start_character":2,"end_line":44,"end_character":8},"updated":"2018-12-13 11:54:33.000000000","message":"Provide","commit_id":"f28dd336c69f3facbee8d4daf8f568276ce707ee"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"a1fae2c34c0c8b7473cea90d96e66215b7628006","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Allow reinitialization of a volume driver when it failed to initialize."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- Import a configuration to set the maximum retry numbers."},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"- About the inveral of retry. The interval of publishing volume driver\u0027s"},{"line_number":47,"context_line":"  capability will be used as the original interval. Later every time if"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_0e0038b0","line":44,"range":{"start_line":44,"start_character":2,"end_line":44,"end_character":8},"in_reply_to":"3f79a3b5_530fb72b","updated":"2018-12-13 12:47:10.000000000","message":"Done","commit_id":"f28dd336c69f3facbee8d4daf8f568276ce707ee"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"bbac26c279b52431b56d6c174170195c46a440d0","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Provide a configuration to set the maximum retry numbers."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- The interval of retry will be exponential backoff."},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"- Retry will be handled in init_host."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_d45f2376","line":44,"updated":"2018-12-13 22:01:43.000000000","message":"-1: I think this needs to be clarified and possibly expanded.\n\nQuestions to answer:\n- What is the initial interval?\n- Is the initial interval hard coded, or should there be a config option?\n- If 0 means no limitation (L52), then should there be an upper limit on the exponential backoff? The backoff algorithm could double the interval until it reaches some max value, after which it stops increasing the interval. I would be OK hard coding the upper interval, but it might just be easier to make it an option.\n\nAn example might also be useful. Given an initial interval of 30 seconds and a \u0027reinit_driver_count\u0027 of 3, the behavior would be as follows:\n\nDelay  Description\n-----   ---------------\n0      First initialization attempt\n30     Reinit attempt #1\n60     Reinit attempt #2\n120    Reinit attempt #3 (last attempt)","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"fe45ef369069efa959aca80bcb018a551e43cd44","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Provide a configuration to set the maximum retry numbers."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- The interval of retry will be exponential backoff."},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"- Retry will be handled in init_host."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_23d651c3","line":44,"in_reply_to":"3f79a3b5_05b1e700","updated":"2018-12-14 14:52:41.000000000","message":"That sounds OK, especially when considering the use case where the storage service is expected to be available \"soon.\" A small number of retries a few seconds apart seems reasonable.","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"28dfd3decf71eeb2b103dca343a1e5bc703a9624","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Provide a configuration to set the maximum retry numbers."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- The interval of retry will be exponential backoff."},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"- Retry will be handled in init_host."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_05b1e700","line":44,"in_reply_to":"3f79a3b5_d45f2376","updated":"2018-12-14 01:06:56.000000000","message":"How about we use similar method as https://github.com/openstack/cinder/blob/master/cinder/volume/driver.py#L452 ? \nThe behavior will be:\nDelay Description\n0 First initialization\n1 reinit #1\n4 reinit #2\n9 reinit #3\n\nMoreover, I would like to remove the value 0. No limitation is not a good choice.\nAs Sean suggested before, it is better to import too many options in config.","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"bbac26c279b52431b56d6c174170195c46a440d0","unresolved":false,"context_lines":[{"line_number":50,"context_line":"- \u0027reinit_driver_count\u0027 (default: 3)"},{"line_number":51,"context_line":"   Set the maximum times to reintialize the driver if volume initialization fails."},{"line_number":52,"context_line":"   Default number is 3. The value 0 means no limitation."},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Alternatives"},{"line_number":55,"context_line":"------------"},{"line_number":56,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_d44883b6","line":53,"updated":"2018-12-13 22:01:43.000000000","message":"Based on my comment above, I\u0027m curious if people think having these additional options would be a good idea:\n\n\u0027reinit_driver_interval\u0027\n\u0027reinit_driver_interval_max\u0027","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"fe45ef369069efa959aca80bcb018a551e43cd44","unresolved":false,"context_lines":[{"line_number":50,"context_line":"- \u0027reinit_driver_count\u0027 (default: 3)"},{"line_number":51,"context_line":"   Set the maximum times to reintialize the driver if volume initialization fails."},{"line_number":52,"context_line":"   Default number is 3. The value 0 means no limitation."},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Alternatives"},{"line_number":55,"context_line":"------------"},{"line_number":56,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_63a3e946","line":53,"in_reply_to":"3f79a3b5_c5c20f56","updated":"2018-12-14 14:52:41.000000000","message":"Yes, I agree there isn\u0027t a good reason to add more options. Thank you for the explanation.","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"28dfd3decf71eeb2b103dca343a1e5bc703a9624","unresolved":false,"context_lines":[{"line_number":50,"context_line":"- \u0027reinit_driver_count\u0027 (default: 3)"},{"line_number":51,"context_line":"   Set the maximum times to reintialize the driver if volume initialization fails."},{"line_number":52,"context_line":"   Default number is 3. The value 0 means no limitation."},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Alternatives"},{"line_number":55,"context_line":"------------"},{"line_number":56,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_c5c20f56","line":53,"in_reply_to":"3f79a3b5_d44883b6","updated":"2018-12-14 01:06:56.000000000","message":"How about the above interval method? I would like to not add too many options into cinder config.","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"bbac26c279b52431b56d6c174170195c46a440d0","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"- We also can differentiate whether it should retry with an exception. Like"},{"line_number":58,"context_line":"  import error, config error, it may not retry. But the benefit is not"},{"line_number":59,"context_line":"  very impressive, and importing the differentiation needs to work in every driver."},{"line_number":60,"context_line":"  As drivers doesn\u0027t differentiate such errors from backend storage errors."},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Data model impact"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_071d6f60","line":59,"range":{"start_line":59,"start_character":23,"end_line":59,"end_character":32},"updated":"2018-12-13 22:01:43.000000000","message":"nit: I think you mean \"... implementing the differentiation...\" (not \"importing\")","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"28dfd3decf71eeb2b103dca343a1e5bc703a9624","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"- We also can differentiate whether it should retry with an exception. Like"},{"line_number":58,"context_line":"  import error, config error, it may not retry. But the benefit is not"},{"line_number":59,"context_line":"  very impressive, and importing the differentiation needs to work in every driver."},{"line_number":60,"context_line":"  As drivers doesn\u0027t differentiate such errors from backend storage errors."},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Data model impact"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_2597434f","line":59,"range":{"start_line":59,"start_character":23,"end_line":59,"end_character":32},"in_reply_to":"3f79a3b5_071d6f60","updated":"2018-12-14 01:06:56.000000000","message":"Done","commit_id":"16437113ef034283c4d8fba0ca33dbd94df7cb56"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"53be4dc3a26dc920af3afe5cdd1a3858d9cf9f30","unresolved":false,"context_lines":[{"line_number":18,"context_line":""},{"line_number":19,"context_line":"During Cinder initialization, for many reasons, the storage backend might not"},{"line_number":20,"context_line":"be ready and responding. In this case, the driver will not be loaded even if"},{"line_number":21,"context_line":"the array gets available right after."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"As there is no retry in Cinder volume service, even later the backend storage"},{"line_number":24,"context_line":"is ready, Cinder volume service can\u0027t recover by itself. It needs users"}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_318193fe","line":21,"range":{"start_line":21,"start_character":10,"end_line":21,"end_character":14},"updated":"2018-12-17 19:37:50.000000000","message":"becomes","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"61f7f79101277d82b495021ca333cd407652f24a","unresolved":false,"context_lines":[{"line_number":18,"context_line":""},{"line_number":19,"context_line":"During Cinder initialization, for many reasons, the storage backend might not"},{"line_number":20,"context_line":"be ready and responding. In this case, the driver will not be loaded even if"},{"line_number":21,"context_line":"the array gets available right after."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"As there is no retry in Cinder volume service, even later the backend storage"},{"line_number":24,"context_line":"is ready, Cinder volume service can\u0027t recover by itself. It needs users"}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_3b73ad1d","line":21,"range":{"start_line":21,"start_character":10,"end_line":21,"end_character":14},"in_reply_to":"3f79a3b5_318193fe","updated":"2018-12-18 01:02:15.000000000","message":"Done","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"53be4dc3a26dc920af3afe5cdd1a3858d9cf9f30","unresolved":false,"context_lines":[{"line_number":27,"context_line":"Use Cases"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":29,"context_line":""},{"line_number":30,"context_line":"When a Cinder volume service starts, sometimes its corresponding storage"},{"line_number":31,"context_line":"services are not ready. But soon the storage services become ready."},{"line_number":32,"context_line":"The users probably prefer Cinder volume services to automatically recover from"},{"line_number":33,"context_line":"the temporary failures without manual intervention of restarting the service."}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_b135a346","line":30,"updated":"2018-12-17 19:37:50.000000000","message":"I think the original use case text was better worded.","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"61f7f79101277d82b495021ca333cd407652f24a","unresolved":false,"context_lines":[{"line_number":27,"context_line":"Use Cases"},{"line_number":28,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":29,"context_line":""},{"line_number":30,"context_line":"When a Cinder volume service starts, sometimes its corresponding storage"},{"line_number":31,"context_line":"services are not ready. But soon the storage services become ready."},{"line_number":32,"context_line":"The users probably prefer Cinder volume services to automatically recover from"},{"line_number":33,"context_line":"the temporary failures without manual intervention of restarting the service."}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_1be7d1b0","line":30,"in_reply_to":"3f79a3b5_b135a346","updated":"2018-12-18 01:02:15.000000000","message":"Done","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":7198,"name":"Jay Bryant","email":"jungleboyj@electronicjungle.net","username":"jsbryant"},"change_message_id":"43d7414c711d1e894fb97732168b3afbce88e213","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Provide a configuration to set the maximum retry numbers."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- The interval of retry will be exponential backoff. Every interval is the"},{"line_number":45,"context_line":"  exponentiation of retry count. The first interval is 1s, second interval"},{"line_number":46,"context_line":"  is 4s, third interval is 9s, and so on."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_6c2c1494","line":44,"range":{"start_line":44,"start_character":28,"end_line":44,"end_character":43},"updated":"2018-12-17 20:07:07.000000000","message":"exponentially backoff","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"61f7f79101277d82b495021ca333cd407652f24a","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"- Provide a configuration to set the maximum retry numbers."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"- The interval of retry will be exponential backoff. Every interval is the"},{"line_number":45,"context_line":"  exponentiation of retry count. The first interval is 1s, second interval"},{"line_number":46,"context_line":"  is 4s, third interval is 9s, and so on."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_9b63e146","line":44,"range":{"start_line":44,"start_character":28,"end_line":44,"end_character":43},"in_reply_to":"3f79a3b5_6c2c1494","updated":"2018-12-18 01:02:15.000000000","message":"Done","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":11904,"name":"Sean McGinnis","email":"sean.mcginnis@gmail.com","username":"SeanM"},"change_message_id":"53be4dc3a26dc920af3afe5cdd1a3858d9cf9f30","unresolved":false,"context_lines":[{"line_number":56,"context_line":"Alternatives"},{"line_number":57,"context_line":"------------"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"- We also can differentiate whether it should retry with an exception. Like"},{"line_number":60,"context_line":"  import error, config error, it may not retry. But the benefit is not"},{"line_number":61,"context_line":"  very impressive, and implementing the differentiation needs work in every"},{"line_number":62,"context_line":"  driver. As drivers doesn\u0027t differentiate such errors from backend storage"}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_f12b1be6","line":59,"range":{"start_line":59,"start_character":0,"end_line":59,"end_character":1},"updated":"2018-12-17 19:37:50.000000000","message":"I don\u0027t think we want this to be formatted as a bullet item.","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":7198,"name":"Jay Bryant","email":"jungleboyj@electronicjungle.net","username":"jsbryant"},"change_message_id":"43d7414c711d1e894fb97732168b3afbce88e213","unresolved":false,"context_lines":[{"line_number":59,"context_line":"- We also can differentiate whether it should retry with an exception. Like"},{"line_number":60,"context_line":"  import error, config error, it may not retry. But the benefit is not"},{"line_number":61,"context_line":"  very impressive, and implementing the differentiation needs work in every"},{"line_number":62,"context_line":"  driver. As drivers doesn\u0027t differentiate such errors from backend storage"},{"line_number":63,"context_line":"  errors."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Data model impact"}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_4c47d852","line":62,"range":{"start_line":62,"start_character":21,"end_line":62,"end_character":28},"updated":"2018-12-17 20:07:07.000000000","message":"don\u0027t","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"61f7f79101277d82b495021ca333cd407652f24a","unresolved":false,"context_lines":[{"line_number":59,"context_line":"- We also can differentiate whether it should retry with an exception. Like"},{"line_number":60,"context_line":"  import error, config error, it may not retry. But the benefit is not"},{"line_number":61,"context_line":"  very impressive, and implementing the differentiation needs work in every"},{"line_number":62,"context_line":"  driver. As drivers doesn\u0027t differentiate such errors from backend storage"},{"line_number":63,"context_line":"  errors."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Data model impact"}],"source_content_type":"text/x-rst","patch_set":4,"id":"3f79a3b5_db52b96a","line":62,"range":{"start_line":62,"start_character":21,"end_line":62,"end_character":28},"in_reply_to":"3f79a3b5_4c47d852","updated":"2018-12-18 01:02:15.000000000","message":"Done","commit_id":"0f23d7e05bfc33b6208fd15fac1acc450f281759"},{"author":{"_account_id":20722,"name":"Yikun Jiang","email":"yikunkero@gmail.com","username":"yikunkero"},"change_message_id":"b9c6964fd9f39ca55c601f2929fe06a24d401dbb","unresolved":false,"context_lines":[{"line_number":44,"context_line":""},{"line_number":45,"context_line":"- The interval of retry will exponentially backoff. Every interval is the"},{"line_number":46,"context_line":"  exponentiation of retry count. The first interval is 1s, second interval"},{"line_number":47,"context_line":"  is 4s, third interval is 9s, and so on."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"- Retry will be handled in init_host."},{"line_number":50,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"3f79a3b5_0143e26f","line":47,"range":{"start_line":47,"start_character":27,"end_line":47,"end_character":29},"updated":"2018-12-18 03:35:21.000000000","message":"https://en.wikipedia.org/wiki/Exponential_backoff\n\n1, 2, 4, 8, 16","commit_id":"824cdc1b5b99fdd7c89813c2f6aa8baadfb86afb"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"317b67359b70954925f588fd4f9f9a6a54fdadd7","unresolved":false,"context_lines":[{"line_number":44,"context_line":""},{"line_number":45,"context_line":"- The interval of retry will exponentially backoff. Every interval is the"},{"line_number":46,"context_line":"  exponentiation of retry count. The first interval is 1s, second interval"},{"line_number":47,"context_line":"  is 4s, third interval is 9s, and so on."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"- Retry will be handled in init_host."},{"line_number":50,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"3f79a3b5_f9fbfac1","line":47,"range":{"start_line":47,"start_character":27,"end_line":47,"end_character":29},"in_reply_to":"3f79a3b5_0143e26f","updated":"2018-12-18 06:31:42.000000000","message":"You are right.","commit_id":"824cdc1b5b99fdd7c89813c2f6aa8baadfb86afb"}]}
