)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"b4df1a8ceac131e3d1c9d50e6450c52b6752d724","unresolved":false,"context_lines":[{"line_number":15,"context_line":"Change-Id: I3b8191f7653d2dccbcb0dd6b45b5b94cb56ba4cf"},{"line_number":16,"context_line":"Story: #2004581"},{"line_number":17,"context_line":"Task: #28366"},{"line_number":18,"context_line":"Depends-On: https://review.openstack.org/#/c/644800/"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":5,"id":"5fc1f717_4a51b735","line":18,"range":{"start_line":18,"start_character":0,"end_line":18,"end_character":52},"updated":"2019-03-27 05:42:10.000000000","message":"This is not required. You can rebase your patch on top of it.","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"}],"specs/approved/software-raid.rst":[{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"856304bc2c00b3b65c9cab9f24b0abed5384bd87","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":3,"id":"5fc1f717_0ea7ec51","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":16},"updated":"2019-03-20 05:01:02.000000000","message":"We need this information only after the RAID is being created successfully. It would be better to set something explicitly through Raid interface (in say node \u0027properties\u0027) to indicate that the RAID configured on the baremetal is a \u0027software\u0027 RAID. Currently, we do set some properties on node[1] upon successful creation of RAID.\n[1] https://github.com/openstack/ironic/blob/master/ironic/common/raid.py#L91-L128","commit_id":"d344ceb8f6721ad1d8cd7db18bb1025122a4dd62"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"db635d27f76ddb98c0dd4992b5bf1b3eda1466c1","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":3,"id":"5fc1f717_b4a38a00","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":16},"in_reply_to":"5fc1f717_0ea7ec51","updated":"2019-03-20 07:51:08.000000000","message":"This field is already used upon creation: to clearly signal that we want a software RAID and to protect any accidental use (since it will be part of the generic h/w manager it is quite exposed).\n\nAre you suggesting to set such a node property none the less?\nIf so, could you please explain what the use case would be?\n\nThe code you linked refers to capabilities for scheduling, but I think all of this has moved to resource classes.","commit_id":"d344ceb8f6721ad1d8cd7db18bb1025122a4dd62"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"682b79f18842eb108ab7cdb052578d561d26fe4a","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":3,"id":"5fc1f717_d49c165f","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":16},"in_reply_to":"5fc1f717_b4a38a00","updated":"2019-03-20 08:05:02.000000000","message":"I was suggesting is using something that would be set by raid.create_configuration() instead of something that is being set by the user. It is possible that \u0027target_raid_config\u0027 is set on the node but \u0027cleaning\u0027 was never invoked. If we set something during create_configuration and remove it when delete_configuration is called, it helps in knowing that raid configuration is indeed present on the bare metal.","commit_id":"d344ceb8f6721ad1d8cd7db18bb1025122a4dd62"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"856304bc2c00b3b65c9cab9f24b0abed5384bd87","unresolved":false,"context_lines":[{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"},{"line_number":93,"context_line":"installed onto the underlying holder disks, and this property will indicate"},{"line_number":94,"context_line":"when to do this."},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"An example of a valid software RAID configuration would hence look like::"},{"line_number":97,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"5fc1f717_4e5bb453","line":94,"range":{"start_line":92,"start_character":39,"end_line":94,"end_character":16},"updated":"2019-03-20 05:01:02.000000000","message":"Can we re-install the bootloader for the whole disk image? Or software RAID can be used only with partition images? It would be helpful to clarify here.","commit_id":"d344ceb8f6721ad1d8cd7db18bb1025122a4dd62"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"db635d27f76ddb98c0dd4992b5bf1b3eda1466c1","unresolved":false,"context_lines":[{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"},{"line_number":93,"context_line":"installed onto the underlying holder disks, and this property will indicate"},{"line_number":94,"context_line":"when to do this."},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"An example of a valid software RAID configuration would hence look like::"},{"line_number":97,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"5fc1f717_54b4a635","line":94,"range":{"start_line":92,"start_character":39,"end_line":94,"end_character":16},"in_reply_to":"5fc1f717_4e5bb453","updated":"2019-03-20 07:51:08.000000000","message":"I only tried with whole-disk images. The current code will certainly not work with partition images (since there is no mechanism to identify which one is the root disk).\n\nIt\u0027s good to clarify this, I\u0027ll update the spec accordingly.","commit_id":"d344ceb8f6721ad1d8cd7db18bb1025122a4dd62"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"b4df1a8ceac131e3d1c9d50e6450c52b6752d724","unresolved":false,"context_lines":[{"line_number":76,"context_line":"more complex RAID configurations."},{"line_number":77,"context_line":""},{"line_number":78,"context_line":"In order to signal that a software RAID configuration is indeed desired (and"},{"line_number":79,"context_line":"to protect from a situation where a software RAID is set up accidentally"},{"line_number":80,"context_line":"when the configuration passed via the ``target_raid_config`` was meant for a"},{"line_number":81,"context_line":"hardware RAID setup, for instance), the ``controller`` property of all of the"},{"line_number":82,"context_line":"logical disks needs to be set to ``software``. Without this setting, the"},{"line_number":83,"context_line":"software RAID code in the GenericHardwareManager of the IPA will ignore the"},{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_aa8a9bbb","line":81,"range":{"start_line":79,"start_character":0,"end_line":81,"end_character":34},"updated":"2019-03-27 05:42:10.000000000","message":"I\u0027m not sure, how this would happen. May be you need to elaborate more here. All the hardware RAID implementation of hardware_manager required to set their \u0027HARDWARE_MANAGER_VERSION\u0027 greater of equal to 3. The Generic(IPA) implementation would set it to not more than 2 [1]. In that sense, if there is a hardware RAID implementation, it would get called before generic implementation. This would be useful if the hardware RAID implementation were to check the controller info in target_raid_config and raise IncompatibleHardwareMethodError if its set to \u0027software\u0027. This would make it invoke IPA implementation of RAID [2] and [3].\nIn that sense, we need to capture it accordingly.\n[1] https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/hardware.py#L249-L263\n[2] https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/extensions/clean.py#L76-L82\n[3] https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/hardware.py#L1252-L1276","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"cbe13b44cbd6302532c1b58efe046e5ce594be1b","unresolved":false,"context_lines":[{"line_number":76,"context_line":"more complex RAID configurations."},{"line_number":77,"context_line":""},{"line_number":78,"context_line":"In order to signal that a software RAID configuration is indeed desired (and"},{"line_number":79,"context_line":"to protect from a situation where a software RAID is set up accidentally"},{"line_number":80,"context_line":"when the configuration passed via the ``target_raid_config`` was meant for a"},{"line_number":81,"context_line":"hardware RAID setup, for instance), the ``controller`` property of all of the"},{"line_number":82,"context_line":"logical disks needs to be set to ``software``. Without this setting, the"},{"line_number":83,"context_line":"software RAID code in the GenericHardwareManager of the IPA will ignore the"},{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_6e66e7ef","line":81,"range":{"start_line":79,"start_character":0,"end_line":81,"end_character":34},"in_reply_to":"5fc1f717_aa8a9bbb","updated":"2019-03-28 11:55:23.000000000","message":"The risk that a s/w RAID setup attempt is made when the config was intended for a custom h/w manager was raised on the channel. It is true that this should not happen as long as everything is configured properly. However, I think it\u0027d be good to require an operator to be specific about what should be set up and have safety measure to prevent a h/w config going into the s/w code of the generic h/w manager and vice versa.\n\nIn any case, the check in the h/w manager is not strictly required it\u0027s just sth we can profit from as we need a flag anyway to trigger the installation of the bootloader.","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"b4df1a8ceac131e3d1c9d50e6450c52b6752d724","unresolved":false,"context_lines":[{"line_number":79,"context_line":"to protect from a situation where a software RAID is set up accidentally"},{"line_number":80,"context_line":"when the configuration passed via the ``target_raid_config`` was meant for a"},{"line_number":81,"context_line":"hardware RAID setup, for instance), the ``controller`` property of all of the"},{"line_number":82,"context_line":"logical disks needs to be set to ``software``. Without this setting, the"},{"line_number":83,"context_line":"software RAID code in the GenericHardwareManager of the IPA will ignore the"},{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_eab5037f","line":85,"range":{"start_line":82,"start_character":47,"end_line":85,"end_character":39},"updated":"2019-03-27 05:42:10.000000000","message":"Wont this break backward compatibility?","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"dddbcbdbeeb8d08beee270bd4a750a9b20b5cf9a","unresolved":false,"context_lines":[{"line_number":79,"context_line":"to protect from a situation where a software RAID is set up accidentally"},{"line_number":80,"context_line":"when the configuration passed via the ``target_raid_config`` was meant for a"},{"line_number":81,"context_line":"hardware RAID setup, for instance), the ``controller`` property of all of the"},{"line_number":82,"context_line":"logical disks needs to be set to ``software``. Without this setting, the"},{"line_number":83,"context_line":"software RAID code in the GenericHardwareManager of the IPA will ignore the"},{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_420f4436","line":85,"range":{"start_line":82,"start_character":47,"end_line":85,"end_character":39},"in_reply_to":"5fc1f717_eab5037f","updated":"2019-03-28 09:01:50.000000000","message":"Ignore this. It is not applicable. I thought we already have software RAID...","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"dddbcbdbeeb8d08beee270bd4a750a9b20b5cf9a","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_2d15493d","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":15},"updated":"2019-03-28 09:01:50.000000000","message":"This data can be returned as part of raid config returned by by the IPA-\u003ecreate_configuration() and update it into node [1] for consumption during deploy.\nhttps://github.com/openstack/ironic/blob/master/ironic/common/raid.py#L91-L128","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"cbe13b44cbd6302532c1b58efe046e5ce594be1b","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_6ed90781","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":15},"in_reply_to":"5fc1f717_2d15493d","updated":"2019-03-28 11:55:23.000000000","message":"How would ironic know when to trigger the bootloader installation?\n\nAlso: rather than updating the node, I\u0027d prefer to use the post create_configuration hook [1]. I\u0027m looking into how to get that working outside of manual cleaning. Seems it requires a get_clean_steps call in AgentRAID ...\n\n[1] https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/agent.py#L755-L783","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":26340,"name":"Ilya Etingof","email":"etingof@gmail.com","username":"etingof"},"change_message_id":"f0008c2173b3c45652e9748d1b736051bcdb67f4","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_b0edf6b8","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":15},"in_reply_to":"5fc1f717_6ed90781","updated":"2019-03-29 13:27:08.000000000","message":"\u003e will also be used by the conductor to identify a software RAID and trigger the required installation of the bootloader.\n\nThis heuristics seems somewhat rigid to me. May be explicit boot loader installation step (which could be enabled/disabled) would work better in the long run?","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"82c8ab9f17702a0f28107ae6c37435ae2f3f27e8","unresolved":false,"context_lines":[{"line_number":84,"context_line":"given ``target_raid_config``. If it is set on only one of the logical drives,"},{"line_number":85,"context_line":"the validation code will raise an error."},{"line_number":86,"context_line":""},{"line_number":87,"context_line":"The ``controller`` property set to ``software`` will also be used by the"},{"line_number":88,"context_line":"conductor to identify a software RAID and trigger the required installation of"},{"line_number":89,"context_line":"the bootloader. While whole-disk images are expected to come with a bootloader"},{"line_number":90,"context_line":"configuration as part of the image, for software RAIDs in the current design"},{"line_number":91,"context_line":"the image will not be at the start of a real disk, but inside the first"},{"line_number":92,"context_line":"partition on top of a software RAID-1. The bootloader must hence be explicitly"}],"source_content_type":"text/x-rst","patch_set":5,"id":"5fc1f717_1306541e","line":89,"range":{"start_line":87,"start_character":0,"end_line":89,"end_character":15},"in_reply_to":"5fc1f717_b0edf6b8","updated":"2019-03-29 13:51:30.000000000","message":"Yes, maybe ... how would such an additional step be triggered/enabled? In the end this is only needed for software RAID, so I guess we\u0027d need some discriminator\nin the end anyway.","commit_id":"6b1a7688fd54e0009c236e9ac23d017a18b9b115"},{"author":{"_account_id":26340,"name":"Ilya Etingof","email":"etingof@gmail.com","username":"etingof"},"change_message_id":"f0008c2173b3c45652e9748d1b736051bcdb67f4","unresolved":false,"context_lines":[{"line_number":99,"context_line":"            {"},{"line_number":100,"context_line":"                \"size_gb\": 100,"},{"line_number":101,"context_line":"                \"raid_level\": \"1\","},{"line_number":102,"context_line":"                \"controller\": \"software\","},{"line_number":103,"context_line":"            },"},{"line_number":104,"context_line":"            {"},{"line_number":105,"context_line":"                \"size_gb\": \"MAX\","}],"source_content_type":"text/x-rst","patch_set":6,"id":"5fc1f717_10816afc","line":102,"range":{"start_line":102,"start_character":40,"end_line":102,"end_character":41},"updated":"2019-03-29 13:27:08.000000000","message":"nit: no comma if it\u0027s intended to be a well-formed JSON?","commit_id":"81134e0de8e694cafa914d17b0baa4f4fec63983"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"82c8ab9f17702a0f28107ae6c37435ae2f3f27e8","unresolved":false,"context_lines":[{"line_number":99,"context_line":"            {"},{"line_number":100,"context_line":"                \"size_gb\": 100,"},{"line_number":101,"context_line":"                \"raid_level\": \"1\","},{"line_number":102,"context_line":"                \"controller\": \"software\","},{"line_number":103,"context_line":"            },"},{"line_number":104,"context_line":"            {"},{"line_number":105,"context_line":"                \"size_gb\": \"MAX\","}],"source_content_type":"text/x-rst","patch_set":6,"id":"5fc1f717_33803047","line":102,"range":{"start_line":102,"start_character":40,"end_line":102,"end_character":41},"in_reply_to":"5fc1f717_10816afc","updated":"2019-03-29 13:51:30.000000000","message":"Done","commit_id":"81134e0de8e694cafa914d17b0baa4f4fec63983"},{"author":{"_account_id":26340,"name":"Ilya Etingof","email":"etingof@gmail.com","username":"etingof"},"change_message_id":"f0008c2173b3c45652e9748d1b736051bcdb67f4","unresolved":false,"context_lines":[{"line_number":104,"context_line":"            {"},{"line_number":105,"context_line":"                \"size_gb\": \"MAX\","},{"line_number":106,"context_line":"                \"raid_level\": \"0\","},{"line_number":107,"context_line":"                \"controller\": \"software\","},{"line_number":108,"context_line":"            }"},{"line_number":109,"context_line":"        ]"},{"line_number":110,"context_line":"    }"}],"source_content_type":"text/x-rst","patch_set":6,"id":"5fc1f717_d07a0224","line":107,"range":{"start_line":107,"start_character":39,"end_line":107,"end_character":41},"updated":"2019-03-29 13:27:08.000000000","message":"ditto","commit_id":"81134e0de8e694cafa914d17b0baa4f4fec63983"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"82c8ab9f17702a0f28107ae6c37435ae2f3f27e8","unresolved":false,"context_lines":[{"line_number":104,"context_line":"            {"},{"line_number":105,"context_line":"                \"size_gb\": \"MAX\","},{"line_number":106,"context_line":"                \"raid_level\": \"0\","},{"line_number":107,"context_line":"                \"controller\": \"software\","},{"line_number":108,"context_line":"            }"},{"line_number":109,"context_line":"        ]"},{"line_number":110,"context_line":"    }"}],"source_content_type":"text/x-rst","patch_set":6,"id":"5fc1f717_d3944c8f","line":107,"range":{"start_line":107,"start_character":39,"end_line":107,"end_character":41},"in_reply_to":"5fc1f717_d07a0224","updated":"2019-03-29 13:51:30.000000000","message":"Done","commit_id":"81134e0de8e694cafa914d17b0baa4f4fec63983"},{"author":{"_account_id":11655,"name":"Julia Kreger","email":"juliaashleykreger@gmail.com","username":"jkreger","status":"Flying to the moon with a Jetpack!"},"change_message_id":"4a7850851a6490bf58bfaa4647eeca7edfd593be","unresolved":false,"context_lines":[{"line_number":116,"context_line":"device are left for follow-up enhancements and beyond the scope of this"},{"line_number":117,"context_line":"specification."},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Also, there is currently no support for partition images, only whole disk"},{"line_number":120,"context_line":"images are supported."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"A first prototype very close to the proposal is available from [2][3][4]."},{"line_number":123,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"5fc1f717_581ffe6f","line":120,"range":{"start_line":119,"start_character":0,"end_line":120,"end_character":21},"updated":"2019-04-01 13:55:43.000000000","message":"Okay, this is confusing. This compared to the sentence on 89-92 are really confusing. I\u0027m not sure what exactly this is referring to.","commit_id":"6053cae0cdd01b141da670921f2205008929b268"},{"author":{"_account_id":11292,"name":"Arne Wiebalck","email":"Arne.Wiebalck@cern.ch","username":"wiebalck"},"change_message_id":"24f66b5a635e4b32e203f67ba4845b0a17a2fc1b","unresolved":false,"context_lines":[{"line_number":116,"context_line":"device are left for follow-up enhancements and beyond the scope of this"},{"line_number":117,"context_line":"specification."},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Also, there is currently no support for partition images, only whole disk"},{"line_number":120,"context_line":"images are supported."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"A first prototype very close to the proposal is available from [2][3][4]."},{"line_number":123,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"5fc1f717_91e1c0ff","line":120,"range":{"start_line":119,"start_character":0,"end_line":120,"end_character":21},"in_reply_to":"5fc1f717_581ffe6f","updated":"2019-04-01 16:03:58.000000000","message":"This is mostly a safety belt to be clear on what has not been considered ... IOW sth I can point to later ;-)\n\nSrsly: partition images need the bootloader to be installed (which ironic does), but I doubt with s/w RAID it would go into the correct place. I have not tested the code with a partition image. This is something that would probably need some attention later on.","commit_id":"6053cae0cdd01b141da670921f2205008929b268"},{"author":{"_account_id":11655,"name":"Julia Kreger","email":"juliaashleykreger@gmail.com","username":"jkreger","status":"Flying to the moon with a Jetpack!"},"change_message_id":"ff49ce50a4a1ffe9fc6f3eeb1bd8eeb432d66d47","unresolved":false,"context_lines":[{"line_number":116,"context_line":"device are left for follow-up enhancements and beyond the scope of this"},{"line_number":117,"context_line":"specification."},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Also, there is currently no support for partition images, only whole disk"},{"line_number":120,"context_line":"images are supported."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"A first prototype very close to the proposal is available from [2][3][4]."},{"line_number":123,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"5fc1f717_a79f5ab2","line":120,"range":{"start_line":119,"start_character":0,"end_line":120,"end_character":21},"in_reply_to":"5fc1f717_91e1c0ff","updated":"2019-04-08 14:30:43.000000000","message":"Okay, this feels like we\u0027re going to need some explicit documentation because this needed because I interpreted the way this would work to be soley with partition images.","commit_id":"6053cae0cdd01b141da670921f2205008929b268"},{"author":{"_account_id":11076,"name":"Shivanand Tendulker","email":"stendulker@gmail.com","username":"stendulker"},"change_message_id":"e4df0e8c6a51d792adeebd4e5aff9f1014081984","unresolved":false,"context_lines":[{"line_number":116,"context_line":"device are left for follow-up enhancements and beyond the scope of this"},{"line_number":117,"context_line":"specification."},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Also, there is currently no support for partition images, only whole disk"},{"line_number":120,"context_line":"images are supported."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"A first prototype very close to the proposal is available from [2][3][4]."},{"line_number":123,"context_line":""}],"source_content_type":"text/x-rst","patch_set":7,"id":"3fce034c_73bd4f67","line":120,"range":{"start_line":119,"start_character":0,"end_line":120,"end_character":21},"in_reply_to":"5fc1f717_a79f5ab2","updated":"2019-04-11 06:39:14.000000000","message":"+1","commit_id":"6053cae0cdd01b141da670921f2205008929b268"}]}
