)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":8122,"name":"Cyril Roelandt","email":"cyril@redhat.com","username":"cyril.roelandt.enovance"},"change_message_id":"7fb492ab0e520464117f1e60245b165e6cc35c88","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"aedac52f_023717d5","updated":"2023-09-11 15:20:32.000000000","message":"Could we mention b1d0feeba4e16046c7840bd4966f31a0e7527e67 (the commit that deprecated this option) in the commit log?\n\nAlso, according to https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/4/html-single/configuration_guide/index , 0 should be the \"no limit\" value:\n\n\"rados_osd_op_timeout\nDescription\nNumber of seconds that RADOS waits for a response from the OSD before returning an error from a RADOS operation. A value of 0 means no limit.","commit_id":"2179cdd7390fabf2cbcf8b30112e29f19d4c5e88"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"eb05755af830f6b3a4e8f4a7287ba5b028b1169b","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"cd2251f5_2d3e025e","in_reply_to":"aedac52f_023717d5","updated":"2023-09-11 15:48:19.000000000","message":"I\u0027ve updated the commit message to include the commit message.\n\nYou are correct about the behavior of timeout\u003d0. In case user explicitly requires rados_connect-timeout\u003d0 then we override client timeouts by 0 so that timeouts are disabled.","commit_id":"2179cdd7390fabf2cbcf8b30112e29f19d4c5e88"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"67183ce6cbe4406f266f00af54ee65677857595c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"32fe084a_f30d596c","updated":"2023-12-07 13:36:22.000000000","message":"@Takashi: I think I understand what the patch is supposed to do, but the glance_store-src-ceph-tempest is still failing with a timeout.  Can you give an explanation of what\u0027s happening there?  I\u0027m having trouble figuring out why the timeout is occurring on a patch that\u0027s specifically designed to address a timeout issue.  (It may have nothing to do with your patch, so feel free to say that.  But I guess we should recheck to get green CI here?)","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9303,"name":"Abhishek Kekane","email":"akekane@redhat.com","username":"abhishekkekane"},"change_message_id":"e3d344c795a6c95ebfc20d6320b1a4ae7411eedd","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"6e269590_d49be2d1","updated":"2024-01-18 17:26:03.000000000","message":"@takashi san, could you please make the suggested changes?","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":8122,"name":"Cyril Roelandt","email":"cyril@redhat.com","username":"cyril.roelandt.enovance"},"change_message_id":"a91bcad57d31e27776e3d9ad8bc4e7eef0ef28ed","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"35197e79_cf5d02ed","updated":"2023-11-29 15:55:33.000000000","message":"I still do not understand what the default value (-1) is supposed to mean. The docs seem to indicate the value should either be \"0\" (no limit) or a strictly positive value (an amount of seconds). Could you point us to the doc that describes negative values?","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"7aaf70b5049c56fbef98be314bae366cd337d20b","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"136d781b_014615e7","updated":"2024-01-18 17:02:52.000000000","message":"Thanks for the explanation, Takashi.  Code LGTM.  See suggestion inline for enhancing the release note.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"bb2219dab267da8a92e2cca41ad1b177a01f0d3c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"25f6be44_8dfb1b9e","updated":"2023-12-13 20:20:30.000000000","message":"This patch lgtm, and follows similar code that cinder has been using for a long time.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9303,"name":"Abhishek Kekane","email":"akekane@redhat.com","username":"abhishekkekane"},"change_message_id":"aa5d400a70310ffc51e3fa173dda31d5140d59a2","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"296a0dcb_414edd2c","updated":"2023-11-27 06:15:54.000000000","message":"recheck","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"b5018f23d6f971f19c5d14ddd0cb34a7be92a502","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"80156669_068bf929","updated":"2023-12-13 08:25:07.000000000","message":"recheck timeout is irrelevant as I explained","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"d3db0f8ad9a9867676566feec6c1eb67dc7b2d70","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"c0d687c7_71f26a4c","in_reply_to":"32fe084a_f30d596c","updated":"2023-12-08 03:46:06.000000000","message":"The timeout in that job is not caused by this change and is happening consistently with the other patches (the job is skipped unless rbd driver is touched so is not blocking all changes)\n\nTimeout is caused by decreased concurrency of tempest, to decrease resource consumption. Please see https://review.opendev.org/c/openstack/glance_store/+/902556","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"f241344dbd7bd735f1e6aff31c4fe8e4d79ae05a","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"0a3c9a8e_a967e8bf","in_reply_to":"35197e79_cf5d02ed","updated":"2023-12-03 06:06:03.000000000","message":"With the proposed implementation, if the option is set to -1 then the store driver does not override timeout[1]. So the values defined in ceph config file, loaded at [2], will be used.\n\nIn case the option is set to 0 then the store driver explicitly override timeouts by 0, so regardless of the timeouts in ceph.conf the client disables timeouts.\n\n[1] https://review.opendev.org/c/openstack/glance_store/+/894514/5/glance_store/_drivers/rbd.py#304\n[2] https://review.opendev.org/c/openstack/glance_store/+/894514/5/glance_store/_drivers/rbd.py#259","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9303,"name":"Abhishek Kekane","email":"akekane@redhat.com","username":"abhishekkekane"},"change_message_id":"86cb983008acd3483dcfcf60ce65e2cad77ba9ec","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"dcc240d7_458a729a","updated":"2024-01-19 04:43:41.000000000","message":"Thank you!!","commit_id":"c197bbdccadf799be2d86da28f1d2302ccf5e1e8"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"1bf922978ac6bea87aa0a50224e89a0da2450927","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"6282a2cd_dc2ce760","updated":"2024-01-19 08:10:22.000000000","message":"recheck timeout is unrelated. we merged https://review.opendev.org/c/openstack/glance_store/+/902556 which increases timeout.","commit_id":"c197bbdccadf799be2d86da28f1d2302ccf5e1e8"},{"author":{"_account_id":9303,"name":"Abhishek Kekane","email":"akekane@redhat.com","username":"abhishekkekane"},"change_message_id":"d197ee549db8fb1cebf540707895d9f968a0729a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"df8d6f87_ece17d27","updated":"2024-01-19 07:27:46.000000000","message":"recheck unrelated timeout","commit_id":"c197bbdccadf799be2d86da28f1d2302ccf5e1e8"}],"releasenotes/notes/fix-rados_connect_timeout-39e5074bc1a3b65b.yaml":[{"author":{"_account_id":8122,"name":"Cyril Roelandt","email":"cyril@redhat.com","username":"cyril.roelandt.enovance"},"change_message_id":"52131ba1ed6362f74f1f57aa9616705a4d4c5718","unresolved":true,"context_lines":[{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    RBD driver now uses the ``rados_connect_timeout`` config option to override"},{"line_number":5,"context_line":"    timeouts of RADOS client. The default value of this option has been changed"},{"line_number":6,"context_line":"    from -1 to 0, and setting this option to 0 disables timeouts even if"},{"line_number":7,"context_line":"    timeouts are set in ceph.conf file."}],"source_content_type":"text/x-yaml","patch_set":2,"id":"a21f4b02_943500a4","line":6,"range":{"start_line":6,"start_character":22,"end_line":6,"end_character":29},"updated":"2023-09-11 15:21:44.000000000","message":"It seems that the patch changes the default value from 0 to -1, not the other way around.","commit_id":"2179cdd7390fabf2cbcf8b30112e29f19d4c5e88"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"eb05755af830f6b3a4e8f4a7287ba5b028b1169b","unresolved":false,"context_lines":[{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    RBD driver now uses the ``rados_connect_timeout`` config option to override"},{"line_number":5,"context_line":"    timeouts of RADOS client. The default value of this option has been changed"},{"line_number":6,"context_line":"    from -1 to 0, and setting this option to 0 disables timeouts even if"},{"line_number":7,"context_line":"    timeouts are set in ceph.conf file."}],"source_content_type":"text/x-yaml","patch_set":2,"id":"913da726_a998a703","line":6,"range":{"start_line":6,"start_character":22,"end_line":6,"end_character":29},"in_reply_to":"a21f4b02_943500a4","updated":"2023-09-11 15:48:19.000000000","message":"Good catch. I\u0027ve fixed it. Also, giving it another thought, now I think upgrade section would be better so I moved the 2nd description to the upgrade section.","commit_id":"2179cdd7390fabf2cbcf8b30112e29f19d4c5e88"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"bb2219dab267da8a92e2cca41ad1b177a01f0d3c","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"8f11528a_b43e9423","updated":"2023-12-13 20:20:30.000000000","message":"Is it worth mentioning in $TBD section that the option is no longer deprecated? It\u0027s sort implied, but faithful readers of past releases may be confused because they recall the option was previously deprecated.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"870c9e716a1878cb421b0e24fa7dd9d98439f219","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"b0494650_91c31a5d","in_reply_to":"8f11528a_b43e9423","updated":"2024-01-19 04:18:54.000000000","message":"This is addressed in the latest version.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"7aaf70b5049c56fbef98be314bae366cd337d20b","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"56cdcb4d_f735b835","line":12,"updated":"2024-01-18 17:02:52.000000000","message":"I agree with Alan, plus, the \u0027upgrade\u0027 section needs to mention the name of the option (since the two notes here will be separated when the release notes are built).\n\nI suggest rewriting the notes like this:\n\n    features:\n      - |\n        RBD driver: the ``rados_connect_timeout`` config option has been\n        un-deprecated and its behavior has been improved.  A value of ``0``\n        is now respected as requesting a timeout of zero, while a value less\n        than zero indicates that glance_store will not set a timeout but\n        instead will use whatever timeouts are set in the Ceph configuration\n        file.\n    \n    upgrade:\n      - |\n        RBD driver: the default value of the ``rados_connect_timeout`` option\n        has been changed from 0 to -1, so that the RBD driver will by default\n        use the timeout values defined in ``ceph.conf``.   Be aware that\n        setting this option to 0 disables timeouts (that is, the RBD driver\n        will make requests with a timeout of zero), thereby overriding any\n        timeouts that are set in the Ceph configuration file.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9303,"name":"Abhishek Kekane","email":"akekane@redhat.com","username":"abhishekkekane"},"change_message_id":"cf78b6c6f6d2d8397fd6c32d8b4daa668367ea03","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"394221cb_eac0b827","line":12,"in_reply_to":"190a04f8_1d4779c4","updated":"2024-01-18 18:25:30.000000000","message":"yes, I have tried restarting glance with setting 0 as client_mount_timeout in ceph.conf and glance service was waiting 25+ minutes to connect to cluster, so I am assuming 0 means it will wait forever.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9816,"name":"Takashi Kajinami","email":"kajinamit@oss.nttdata.com","username":"kajinamit"},"change_message_id":"870c9e716a1878cb421b0e24fa7dd9d98439f219","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"ed3a37a8_9701e6a2","line":12,"in_reply_to":"394221cb_eac0b827","updated":"2024-01-19 04:18:54.000000000","message":"Thanks for these feedback ! I\u0027ve updated the release note and add description about \u0027never timeout\u0027 behavior.\n\nJust leaving this as a note. Alternatively we may use 0 to use the ceph.conf values and 1 to disable timeout, but I\u0027d prefer the current proposed behavior to make the meaning consistent with cinder.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":9303,"name":"Abhishek Kekane","email":"akekane@redhat.com","username":"abhishekkekane"},"change_message_id":"50e7896f73b22707b5003b72d2ec19edb6b75f6a","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"b4e5ea26_1f85c19d","line":12,"in_reply_to":"56cdcb4d_f735b835","updated":"2024-01-18 17:29:12.000000000","message":"Is it worth mentioning that setting it to 0 means it will never timeouts and always wait for connection?","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"},{"author":{"_account_id":5314,"name":"Brian Rosmaita","email":"rosmaita.fossdev@gmail.com","username":"brian-rosmaita"},"change_message_id":"93095ae6dfb69d0f336ab4cd6718f4be35d49cd0","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":5,"id":"190a04f8_1d4779c4","line":12,"in_reply_to":"b4e5ea26_1f85c19d","updated":"2024-01-18 18:12:10.000000000","message":"Yes, we should definitely mention that!  I was taking 0 to mean \"never wait\", but if RBD takes 0 to mean \"wait forever\", we should definitely point that out.","commit_id":"ff6466dc1df24d50cf16d8e3b456955b2d533299"}]}