)]}'
{"specs/xena/whole-disk-default.rst":[{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"51639ca71f6f939977fe360bf12fe3a4b293b5c5","unresolved":true,"context_lines":[{"line_number":15,"context_line":""},{"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":"The capability to build and deploy a whole-disk overcloud image has been"},{"line_number":20,"context_line":"available for many releases, but it is time to switch to this as the default."},{"line_number":21,"context_line":"Doing this will avoid the following issues and bring the following benefits:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"d9724a47_70c60497","line":18,"updated":"2021-05-04 08:08:49.000000000","message":"please consider adding some words about what \u0027whole-disk\u0027 image is. e.g. \u0027currently TripleO deployments use overcloud-full.qcow2 (partition image) + overcloud-full.initrd + overcloud-full.vmlinuz but we instead want to switch to .. \" etc","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"4d00421f29c50eede541b89a48e0dabfbbac73ed","unresolved":true,"context_lines":[{"line_number":15,"context_line":""},{"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":"The capability to build and deploy a whole-disk overcloud image has been"},{"line_number":20,"context_line":"available for many releases, but it is time to switch to this as the default."},{"line_number":21,"context_line":"Doing this will avoid the following issues and bring the following benefits:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"53c6ec7d_620e974a","line":18,"in_reply_to":"d9724a47_70c60497","updated":"2021-05-05 22:14:55.000000000","message":"I\u0027ve added an explainer section which describes what each image type is, what IPA does when each image type is deployed, and what happens on first boot.","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"51639ca71f6f939977fe360bf12fe3a4b293b5c5","unresolved":true,"context_lines":[{"line_number":39,"context_line":""},{"line_number":40,"context_line":"* The `requirements for a hardened image`_ includes having separate volumes for"},{"line_number":41,"context_line":"  root, data etc. All TripleO users get the security benefit of hardened images"},{"line_number":42,"context_line":"  when a whole-disk image is used."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"Proposed Change"},{"line_number":45,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"79d6ca2c_49b5ac2d","line":42,"updated":"2021-05-04 08:08:49.000000000","message":"Another bullet point here is:\n\n* We currently need dedicated CI jobs both in the upstream check/gate (when the relevant files changed) but also in periodic integration lines, to build and publish the latest \u0027current-tripleo\u0027 version of the hardened images. If we are removing the separate hardened images then we can also reduce our CI footprint by removing those jobs.","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"4d00421f29c50eede541b89a48e0dabfbbac73ed","unresolved":false,"context_lines":[{"line_number":39,"context_line":""},{"line_number":40,"context_line":"* The `requirements for a hardened image`_ includes having separate volumes for"},{"line_number":41,"context_line":"  root, data etc. All TripleO users get the security benefit of hardened images"},{"line_number":42,"context_line":"  when a whole-disk image is used."},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"Proposed Change"},{"line_number":45,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"1754d467_22e4d01d","line":42,"in_reply_to":"79d6ca2c_49b5ac2d","updated":"2021-05-05 22:14:55.000000000","message":"Done","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"51639ca71f6f939977fe360bf12fe3a4b293b5c5","unresolved":true,"context_lines":[{"line_number":68,"context_line":"overcloud-hardened-full can be deleted, leaving only"},{"line_number":69,"context_line":"overcloud-hardened-uefi-full. Other refactoring can be done such as renaming"},{"line_number":70,"context_line":"-python3.yaml back to -base.yaml."},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"Reviewing partition layout"},{"line_number":73,"context_line":"++++++++++++++++++++++++++"},{"line_number":74,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"4295ecf7_17b15934","line":71,"updated":"2021-05-04 08:08:49.000000000","message":"As part of this you can also remove the CI jobs upstream check/gate and also periodic jobs","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"4d00421f29c50eede541b89a48e0dabfbbac73ed","unresolved":true,"context_lines":[{"line_number":68,"context_line":"overcloud-hardened-full can be deleted, leaving only"},{"line_number":69,"context_line":"overcloud-hardened-uefi-full. Other refactoring can be done such as renaming"},{"line_number":70,"context_line":"-python3.yaml back to -base.yaml."},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"Reviewing partition layout"},{"line_number":73,"context_line":"++++++++++++++++++++++++++"},{"line_number":74,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"21e0cf4d_ca15fd17","line":71,"in_reply_to":"4295ecf7_17b15934","updated":"2021-05-05 22:14:55.000000000","message":"I cover this in the \u0027Image building and publishing\u0027 section line #234","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"51639ca71f6f939977fe360bf12fe3a4b293b5c5","unresolved":true,"context_lines":[{"line_number":132,"context_line":"Upgrade Impact"},{"line_number":133,"context_line":"--------------"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":"Nodes upgraded in-place will continue to be partition image based, and"},{"line_number":136,"context_line":"new/replaced nodes will be deployed with whole-disk images. This doesn\u0027t have"},{"line_number":137,"context_line":"a specific upgrade implication, unless we document an option for replacing"},{"line_number":138,"context_line":"every node in order to ensure all nodes are deployed with whole-disk images."}],"source_content_type":"text/x-rst","patch_set":3,"id":"4b6f4977_7f34144f","line":135,"range":{"start_line":135,"start_character":0,"end_line":135,"end_character":53},"updated":"2021-05-04 08:08:49.000000000","message":"hmmm.. so the \u0027remove the CI jobs\u0027 has a caveat in that we will only do that for master onwards we will still need to build the \u0027old\u0027 images including the hardened for anything stable/wallaby and older","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"4d00421f29c50eede541b89a48e0dabfbbac73ed","unresolved":true,"context_lines":[{"line_number":132,"context_line":"Upgrade Impact"},{"line_number":133,"context_line":"--------------"},{"line_number":134,"context_line":""},{"line_number":135,"context_line":"Nodes upgraded in-place will continue to be partition image based, and"},{"line_number":136,"context_line":"new/replaced nodes will be deployed with whole-disk images. This doesn\u0027t have"},{"line_number":137,"context_line":"a specific upgrade implication, unless we document an option for replacing"},{"line_number":138,"context_line":"every node in order to ensure all nodes are deployed with whole-disk images."}],"source_content_type":"text/x-rst","patch_set":3,"id":"f6d4462c_996a8089","line":135,"range":{"start_line":135,"start_character":0,"end_line":135,"end_character":53},"in_reply_to":"4b6f4977_7f34144f","updated":"2021-05-05 22:14:55.000000000","message":"Correct, although I\u0027d propose we backport to wallaby so the old jobs will remain up until victoria. Maybe backporting changes which delete image yaml is too much for wallaby, what do you think?","commit_id":"fde61f79c86b52c2970a802f5068ccb85f9a66db"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"c714a0dc26cbb2ca6cad8d1950e813c6bc98f265","unresolved":true,"context_lines":[{"line_number":128,"context_line":""},{"line_number":129,"context_line":"On node first boot, a replacement mechanism for growing the root partition is"},{"line_number":130,"context_line":"required. This is a harder problem for the multiple LVM volumes which the"},{"line_number":131,"context_line":"whole-disk image creates. Generally the ``/var`` volume should grow to take"},{"line_number":132,"context_line":"available disk space, but sometimes ``/srv`` will need to grow, and sometimes"},{"line_number":133,"context_line":"there may need to be a proportional split of multiple volumes. This suggests"},{"line_number":134,"context_line":"that there will be new tripleo-heat-templates variables which will specify"}],"source_content_type":"text/x-rst","patch_set":4,"id":"65be7515_edfa83e7","line":131,"range":{"start_line":131,"start_character":42,"end_line":131,"end_character":45},"updated":"2021-05-10 08:57:08.000000000","message":"\"because this is where tripleo stores config data for deployment (eg. container files/config)","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"fb878b33c672480ada8e6b58ecdbaf1a48c3a9ff","unresolved":false,"context_lines":[{"line_number":128,"context_line":""},{"line_number":129,"context_line":"On node first boot, a replacement mechanism for growing the root partition is"},{"line_number":130,"context_line":"required. This is a harder problem for the multiple LVM volumes which the"},{"line_number":131,"context_line":"whole-disk image creates. Generally the ``/var`` volume should grow to take"},{"line_number":132,"context_line":"available disk space, but sometimes ``/srv`` will need to grow, and sometimes"},{"line_number":133,"context_line":"there may need to be a proportional split of multiple volumes. This suggests"},{"line_number":134,"context_line":"that there will be new tripleo-heat-templates variables which will specify"}],"source_content_type":"text/x-rst","patch_set":4,"id":"70f0020d_74b40208","line":131,"range":{"start_line":131,"start_character":42,"end_line":131,"end_character":45},"in_reply_to":"65be7515_edfa83e7","updated":"2021-05-18 00:46:42.000000000","message":"Done","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":24245,"name":"Harald Jensås","email":"hjensas@redhat.com","username":"harald.jensas"},"change_message_id":"7eba91656148068f651f2cd1c9db2f77f277107b","unresolved":true,"context_lines":[{"line_number":131,"context_line":"whole-disk image creates. Generally the ``/var`` volume should grow to take"},{"line_number":132,"context_line":"available disk space, but sometimes ``/srv`` will need to grow, and sometimes"},{"line_number":133,"context_line":"there may need to be a proportional split of multiple volumes. This suggests"},{"line_number":134,"context_line":"that there will be new tripleo-heat-templates variables which will specify"},{"line_number":135,"context_line":"the volume/proportion growth behaviour on a per-role basis."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"A new utility is required which automates this LVM volume growing"}],"source_content_type":"text/x-rst","patch_set":4,"id":"59a086b9_c5e02c42","line":134,"range":{"start_line":134,"start_character":23,"end_line":134,"end_character":45},"updated":"2021-05-10 21:58:18.000000000","message":"Should this be a baremetal deployment definition variable?\nOr do we want to develop this as a composable service in THT which is applied by the deploy-steps playbook?","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"4188005e0757ab46a5edbcc2b0e1c9d1c1455840","unresolved":true,"context_lines":[{"line_number":131,"context_line":"whole-disk image creates. Generally the ``/var`` volume should grow to take"},{"line_number":132,"context_line":"available disk space, but sometimes ``/srv`` will need to grow, and sometimes"},{"line_number":133,"context_line":"there may need to be a proportional split of multiple volumes. This suggests"},{"line_number":134,"context_line":"that there will be new tripleo-heat-templates variables which will specify"},{"line_number":135,"context_line":"the volume/proportion growth behaviour on a per-role basis."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"A new utility is required which automates this LVM volume growing"}],"source_content_type":"text/x-rst","patch_set":4,"id":"b716cbb5_8b097de6","line":134,"range":{"start_line":134,"start_character":23,"end_line":134,"end_character":45},"in_reply_to":"59a086b9_c5e02c42","updated":"2021-05-11 00:08:09.000000000","message":"I was thinking as a composable service applied by the deploy-steps playbook. The only reason for this is to provide immediate root-cause feedback (and forced failure) if the grow operation fails. If it fails during cloud-init there will be no feedback to the user until a volume becomes full at some point in the future.","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"616c26e872787cfc593bfbb82d7387846f552376","unresolved":false,"context_lines":[{"line_number":131,"context_line":"whole-disk image creates. Generally the ``/var`` volume should grow to take"},{"line_number":132,"context_line":"available disk space, but sometimes ``/srv`` will need to grow, and sometimes"},{"line_number":133,"context_line":"there may need to be a proportional split of multiple volumes. This suggests"},{"line_number":134,"context_line":"that there will be new tripleo-heat-templates variables which will specify"},{"line_number":135,"context_line":"the volume/proportion growth behaviour on a per-role basis."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"A new utility is required which automates this LVM volume growing"}],"source_content_type":"text/x-rst","patch_set":4,"id":"dc0cae2f_abb0788d","line":134,"range":{"start_line":134,"start_character":23,"end_line":134,"end_character":45},"in_reply_to":"92033e57_944c8871","updated":"2021-05-12 20:44:42.000000000","message":"Thats a good point. I\u0027ll clarify that it should be run by early ansible and between provision and deploy is an option.\n\nThis won\u0027t be backported earlier that Wallaby. For users building their own images there will be a diskimage-builder option to run on first boot, baked into the image. So they can specify to grow /var or whatever at image build time.","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":24245,"name":"Harald Jensås","email":"hjensas@redhat.com","username":"harald.jensas"},"change_message_id":"baa5c49209c4083dab611acff9113c2df1494fad","unresolved":false,"context_lines":[{"line_number":131,"context_line":"whole-disk image creates. Generally the ``/var`` volume should grow to take"},{"line_number":132,"context_line":"available disk space, but sometimes ``/srv`` will need to grow, and sometimes"},{"line_number":133,"context_line":"there may need to be a proportional split of multiple volumes. This suggests"},{"line_number":134,"context_line":"that there will be new tripleo-heat-templates variables which will specify"},{"line_number":135,"context_line":"the volume/proportion growth behaviour on a per-role basis."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"A new utility is required which automates this LVM volume growing"}],"source_content_type":"text/x-rst","patch_set":4,"id":"92033e57_944c8871","line":134,"range":{"start_line":134,"start_character":23,"end_line":134,"end_character":45},"in_reply_to":"b716cbb5_8b097de6","updated":"2021-05-12 14:32:40.000000000","message":"With the network-v2 work we have an inventory and can apply an ansible playbook/role post node provisioning, prior to deploying the heat stack.\n\nBut this may need to be back ported to train as well I assume, so a service applied by the deploy-steps playbook is probably better.","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"c714a0dc26cbb2ca6cad8d1950e813c6bc98f265","unresolved":true,"context_lines":[{"line_number":282,"context_line":"overcloud-hardened-uefi-full.qcow2 except that it only supports legacy BIOS"},{"line_number":283,"context_line":"booting. Since overcloud-hardened-uefi-full.qcow2 supports both legacy BIOS"},{"line_number":284,"context_line":"and UEFI boot, the periodic jobs which build overcloud-hardened-full.qcow2"},{"line_number":285,"context_line":"can be switched off."},{"line_number":286,"context_line":""},{"line_number":287,"context_line":"CI support"},{"line_number":288,"context_line":"----------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"ec7f366b_840b1e6f","line":285,"updated":"2021-05-10 08:57:08.000000000","message":"\"at least for xena and wallaby branches (To be determined if we will backport further)\"","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"fb878b33c672480ada8e6b58ecdbaf1a48c3a9ff","unresolved":false,"context_lines":[{"line_number":282,"context_line":"overcloud-hardened-uefi-full.qcow2 except that it only supports legacy BIOS"},{"line_number":283,"context_line":"booting. Since overcloud-hardened-uefi-full.qcow2 supports both legacy BIOS"},{"line_number":284,"context_line":"and UEFI boot, the periodic jobs which build overcloud-hardened-full.qcow2"},{"line_number":285,"context_line":"can be switched off."},{"line_number":286,"context_line":""},{"line_number":287,"context_line":"CI support"},{"line_number":288,"context_line":"----------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"69f4b7d6_c238946b","line":285,"in_reply_to":"ec7f366b_840b1e6f","updated":"2021-05-18 00:46:42.000000000","message":"Done","commit_id":"f86c4f6a965e6e82884430c3d42b3e8baf23980b"},{"author":{"_account_id":8449,"name":"Marios Andreou","email":"marios.andreou@gmail.com","username":"marios"},"change_message_id":"d6711c9bc90d8bbb553b0e9f0bd088c9529ea2b3","unresolved":true,"context_lines":[{"line_number":50,"context_line":"When the partition image deploy boots for the first time, the root partition"},{"line_number":51,"context_line":"grows to take up all of the available disk space. This mechanism is provided"},{"line_number":52,"context_line":"by the base cloud image. There is no equivalent partition growing mechanism"},{"line_number":53,"context_line":"for a multi-volume LVM whole-disk image."},{"line_number":54,"context_line":""},{"line_number":55,"context_line":"Problem Description"},{"line_number":56,"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":5,"id":"83a8c621_ff41a934","line":53,"range":{"start_line":53,"start_character":39,"end_line":53,"end_character":40},"updated":"2021-05-18 06:53:49.000000000","message":"(but will be needed as part of this work) is implied here?","commit_id":"e80d5821515ff89863e2e5275bf2add210520533"},{"author":{"_account_id":4571,"name":"Steve Baker","email":"sbaker@redhat.com","username":"steve-stevebaker"},"change_message_id":"eec4c34db13b518ff32738eb33aff2636956d15b","unresolved":true,"context_lines":[{"line_number":50,"context_line":"When the partition image deploy boots for the first time, the root partition"},{"line_number":51,"context_line":"grows to take up all of the available disk space. This mechanism is provided"},{"line_number":52,"context_line":"by the base cloud image. There is no equivalent partition growing mechanism"},{"line_number":53,"context_line":"for a multi-volume LVM whole-disk image."},{"line_number":54,"context_line":""},{"line_number":55,"context_line":"Problem Description"},{"line_number":56,"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":5,"id":"f7adb1b1_e16daedc","line":53,"range":{"start_line":53,"start_character":39,"end_line":53,"end_character":40},"in_reply_to":"83a8c621_ff41a934","updated":"2021-05-18 22:22:55.000000000","message":"Yes it is needed as part of this work. I\u0027ve got something working already here https://review.opendev.org/c/openstack/diskimage-builder/+/791083","commit_id":"e80d5821515ff89863e2e5275bf2add210520533"}]}