)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"42049d8527c9a4fcff59050f78aa477e9cc5f804","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"859a1a6e_b8f3beec","updated":"2023-05-29 11:08:52.000000000","message":"some minor issues inline but overall this looks good.","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"14aa4f72_82d404f9","updated":"2023-06-08 09:57:01.000000000","message":"First glance looks good with comments, but I still need to sit down correctly and think about all the details and their implications.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"372f5b6fcfea37784426f2ea1263d41edbc72639","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"4aa3f1ef_64bcf7be","updated":"2023-06-08 11:17:17.000000000","message":"just responed to some of sylvains comments\nim still generally ok with this but i also have not deaply review the latest version.\n\nill do a pass of this again when you have adressed sylvains comments.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"b4c02984fdf574175de7877a935a92eb1b876949","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"2e1fd191_8146b90f","updated":"2023-06-13 23:36:37.000000000","message":"I think we can move forward with this proposal with the limited scope. Rejecting the complex case synchronously is hard so either we should simply make it fail during scheduling or implement the scheduler filter that can correlate the two set of request groups (could be hard?!)","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"ab4a4be32ba8e5696405bbbe0417e01ba15714f3","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"a3c7388a_364f6d5b","updated":"2023-06-30 12:02:02.000000000","message":"I think we are still not clear on how the interface attach will be rejected with it would require both a PCI device and QoS resources.","commit_id":"5f3d549cfbc23b7490a94c326d6ea29cebfd867c"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"2cd3bb64aa9717a902c543941e3581a33fc79167","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"a852c9b6_fea2e403","updated":"2023-07-06 16:24:47.000000000","message":"Thanks for writing this spec but as gibi said, you need to write a bit more how you will work on it. \n\nGiven today is the Spec Approval Freeze, my humble opinion is that you should rather focus on working on the implementation changes so next cycle you could provide a new spec explaining what you did and you could provide the links to the code.\nOf course, we can\u0027t merge the code by Bobcat because of the spec freeze, but at least you don\u0027t need to wait for the spec before working on it 😊","commit_id":"5f3d549cfbc23b7490a94c326d6ea29cebfd867c"}],"specs/2023.2/approved/track-sriov-nics-in-placement.rst":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"42049d8527c9a4fcff59050f78aa477e9cc5f804","unresolved":true,"context_lines":[{"line_number":167,"context_line":"of this spec but is intended to be suported in the future."},{"line_number":168,"context_line":""},{"line_number":169,"context_line":"Nova will detect and refuse to boot an instance with an SR-IOV type port that"},{"line_number":170,"context_line":"contains addtional resouce requests. This will be done by returning a code"},{"line_number":171,"context_line":"409 until support for this is added. Attaching SR-IOV ports with additional"},{"line_number":172,"context_line":"resources will also be detected and rejected."},{"line_number":173,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"83fdc53e_e0e73314","line":170,"updated":"2023-05-29 11:08:52.000000000","message":"we should clarify that this only applies to addtional resouce requestr on the port and only if pci in palcement is enabled.","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"f97929954894a911298d11c1c5fbdcb848c7837e","unresolved":false,"context_lines":[{"line_number":167,"context_line":"of this spec but is intended to be suported in the future."},{"line_number":168,"context_line":""},{"line_number":169,"context_line":"Nova will detect and refuse to boot an instance with an SR-IOV type port that"},{"line_number":170,"context_line":"contains addtional resouce requests. This will be done by returning a code"},{"line_number":171,"context_line":"409 until support for this is added. Attaching SR-IOV ports with additional"},{"line_number":172,"context_line":"resources will also be detected and rejected."},{"line_number":173,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"ab84830f_61ee8cf5","line":170,"in_reply_to":"83fdc53e_e0e73314","updated":"2023-05-30 04:55:53.000000000","message":"Done","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"42049d8527c9a4fcff59050f78aa477e9cc5f804","unresolved":true,"context_lines":[{"line_number":170,"context_line":"contains addtional resouce requests. This will be done by returning a code"},{"line_number":171,"context_line":"409 until support for this is added. Attaching SR-IOV ports with additional"},{"line_number":172,"context_line":"resources will also be detected and rejected."},{"line_number":173,"context_line":""},{"line_number":174,"context_line":""},{"line_number":175,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"},{"line_number":176,"context_line":"--------------------------------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"c4612549_c765567e","line":173,"updated":"2023-05-29 11:08:52.000000000","message":"we should also add\n\nphysical_network as a trait to the RP i.e.  CUSTOM_PHYSNET_\u003cphysical_network\u003e","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"f97929954894a911298d11c1c5fbdcb848c7837e","unresolved":false,"context_lines":[{"line_number":170,"context_line":"contains addtional resouce requests. This will be done by returning a code"},{"line_number":171,"context_line":"409 until support for this is added. Attaching SR-IOV ports with additional"},{"line_number":172,"context_line":"resources will also be detected and rejected."},{"line_number":173,"context_line":""},{"line_number":174,"context_line":""},{"line_number":175,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"},{"line_number":176,"context_line":"--------------------------------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"490edbaf_638a2047","line":173,"in_reply_to":"c4612549_c765567e","updated":"2023-05-30 04:55:53.000000000","message":"Done","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"42049d8527c9a4fcff59050f78aa477e9cc5f804","unresolved":true,"context_lines":[{"line_number":290,"context_line":""},{"line_number":291,"context_line":"To utilize the new feature the operator will have to define two new config"},{"line_number":292,"context_line":"options. One to enable the placement scheduling logic and a second to enable"},{"line_number":293,"context_line":"the reporting of the PCI devices to Placement."},{"line_number":294,"context_line":""},{"line_number":295,"context_line":"Developer impact"},{"line_number":296,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"5f821dba_e6d3674c","line":293,"updated":"2023-05-29 11:08:52.000000000","message":"the former already exists as https://docs.openstack.org/nova/latest/configuration/config.html#filter_scheduler.pci_in_placement and the latter is https://docs.openstack.org/nova/latest/configuration/config.html#pci.report_in_placement\n\nno new config options are requried but they will have to opt into this behviaor  by setting the existing options.","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"f97929954894a911298d11c1c5fbdcb848c7837e","unresolved":false,"context_lines":[{"line_number":290,"context_line":""},{"line_number":291,"context_line":"To utilize the new feature the operator will have to define two new config"},{"line_number":292,"context_line":"options. One to enable the placement scheduling logic and a second to enable"},{"line_number":293,"context_line":"the reporting of the PCI devices to Placement."},{"line_number":294,"context_line":""},{"line_number":295,"context_line":"Developer impact"},{"line_number":296,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"dd26f77b_1f405d35","line":293,"in_reply_to":"5f821dba_e6d3674c","updated":"2023-05-30 04:55:53.000000000","message":"Done","commit_id":"8a928a40e11c3131d4fd9c3e71dc958014e2408b"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":true,"context_lines":[{"line_number":13,"context_line":"In the zed and 2023.1 (antelope) releases support was added for tracking"},{"line_number":14,"context_line":"PCI devices that did not contain the physical_network tag in Placement."},{"line_number":15,"context_line":"This enables generic PCI devices that are consumed via flavor based PCI"},{"line_number":16,"context_line":"passthough to be tracked in Placement. PCI devices that are consomed"},{"line_number":17,"context_line":"via Neutron port however are not tracked in Placement. This spec aims to"},{"line_number":18,"context_line":"address that gap and enable tracking of Neutron manged PCI devices."},{"line_number":19,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"99b84a9d_89812e14","line":16,"range":{"start_line":16,"start_character":60,"end_line":16,"end_character":68},"updated":"2023-06-08 09:57:01.000000000","message":"nit: consumed","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"6a5a48d4e283dab455c5c4f7352f7161cbc3f2c5","unresolved":false,"context_lines":[{"line_number":13,"context_line":"In the zed and 2023.1 (antelope) releases support was added for tracking"},{"line_number":14,"context_line":"PCI devices that did not contain the physical_network tag in Placement."},{"line_number":15,"context_line":"This enables generic PCI devices that are consumed via flavor based PCI"},{"line_number":16,"context_line":"passthough to be tracked in Placement. PCI devices that are consomed"},{"line_number":17,"context_line":"via Neutron port however are not tracked in Placement. This spec aims to"},{"line_number":18,"context_line":"address that gap and enable tracking of Neutron manged PCI devices."},{"line_number":19,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3255731a_88474445","line":16,"range":{"start_line":16,"start_character":60,"end_line":16,"end_character":68},"in_reply_to":"99b84a9d_89812e14","updated":"2023-06-09 05:28:28.000000000","message":"Done","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":true,"context_lines":[{"line_number":23,"context_line":""},{"line_number":24,"context_line":"Nova has supported generic stateless PCI passthrough for many releases using a"},{"line_number":25,"context_line":"dedicated PCI tracker in conjunction with a ``PciPassthroughFilter`` scheduler"},{"line_number":26,"context_line":"post filter."},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"The PCI tracker is responsible for tracking which PCI devices are available,"},{"line_number":29,"context_line":"claimed, and allocated, the capabilities of the device, its consumer when"}],"source_content_type":"text/x-rst","patch_set":3,"id":"66042719_08e4115d","line":26,"range":{"start_line":26,"start_character":0,"end_line":26,"end_character":4},"updated":"2023-06-08 09:57:01.000000000","message":"nit: technically, the scheduler predates the tracker for claiming so you can remove this word :)","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"6a5a48d4e283dab455c5c4f7352f7161cbc3f2c5","unresolved":false,"context_lines":[{"line_number":23,"context_line":""},{"line_number":24,"context_line":"Nova has supported generic stateless PCI passthrough for many releases using a"},{"line_number":25,"context_line":"dedicated PCI tracker in conjunction with a ``PciPassthroughFilter`` scheduler"},{"line_number":26,"context_line":"post filter."},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"The PCI tracker is responsible for tracking which PCI devices are available,"},{"line_number":29,"context_line":"claimed, and allocated, the capabilities of the device, its consumer when"}],"source_content_type":"text/x-rst","patch_set":3,"id":"d760cb11_8e59d453","line":26,"range":{"start_line":26,"start_character":0,"end_line":26,"end_character":4},"in_reply_to":"56079d3b_f6fb9cb8","updated":"2023-06-09 05:28:28.000000000","message":"Ack","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"372f5b6fcfea37784426f2ea1263d41edbc72639","unresolved":true,"context_lines":[{"line_number":23,"context_line":""},{"line_number":24,"context_line":"Nova has supported generic stateless PCI passthrough for many releases using a"},{"line_number":25,"context_line":"dedicated PCI tracker in conjunction with a ``PciPassthroughFilter`` scheduler"},{"line_number":26,"context_line":"post filter."},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"The PCI tracker is responsible for tracking which PCI devices are available,"},{"line_number":29,"context_line":"claimed, and allocated, the capabilities of the device, its consumer when"}],"source_content_type":"text/x-rst","patch_set":3,"id":"56079d3b_f6fb9cb8","line":26,"range":{"start_line":26,"start_character":0,"end_line":26,"end_character":4},"in_reply_to":"66042719_08e4115d","updated":"2023-06-08 11:17:17.000000000","message":"the shcudler never claims pci devices\n\nthey are claimed in the pci tracker on the compute.\n\nso this is not wrong also claimed and allocated had a speciic meaning in the context of the pci tracker before placement was created to this statement is correct as is.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":true,"context_lines":[{"line_number":37,"context_line":""},{"line_number":38,"context_line":".. _`flavor extra specs`: https://docs.openstack.org/nova/latest/configuration/extra-specs.html#pci_passthrough:alias"},{"line_number":39,"context_line":""},{"line_number":40,"context_line":"Currently Nova has the capability to model the availablity of flavor managed"},{"line_number":41,"context_line":"PCI devices in Placement but lack the same capability for devices consumed via"},{"line_number":42,"context_line":"Neutron ports. All instance requests or VM with SR-IOV, VDPA, hardware"},{"line_number":43,"context_line":"offloaded OVS or DPU ports rely on the ``PciPassthroughFilter`` to select"}],"source_content_type":"text/x-rst","patch_set":3,"id":"9dff153d_5d3bebba","line":40,"range":{"start_line":40,"start_character":47,"end_line":40,"end_character":58},"updated":"2023-06-08 09:57:01.000000000","message":"nit: availability","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"6a5a48d4e283dab455c5c4f7352f7161cbc3f2c5","unresolved":false,"context_lines":[{"line_number":37,"context_line":""},{"line_number":38,"context_line":".. _`flavor extra specs`: https://docs.openstack.org/nova/latest/configuration/extra-specs.html#pci_passthrough:alias"},{"line_number":39,"context_line":""},{"line_number":40,"context_line":"Currently Nova has the capability to model the availablity of flavor managed"},{"line_number":41,"context_line":"PCI devices in Placement but lack the same capability for devices consumed via"},{"line_number":42,"context_line":"Neutron ports. All instance requests or VM with SR-IOV, VDPA, hardware"},{"line_number":43,"context_line":"offloaded OVS or DPU ports rely on the ``PciPassthroughFilter`` to select"}],"source_content_type":"text/x-rst","patch_set":3,"id":"c3b686cd_33f24576","line":40,"range":{"start_line":40,"start_character":47,"end_line":40,"end_character":58},"in_reply_to":"9dff153d_5d3bebba","updated":"2023-06-09 05:28:28.000000000","message":"Done","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":true,"context_lines":[{"line_number":46,"context_line":""},{"line_number":47,"context_line":".. rubric:: Limitations"},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"* The current implementation is functionally slow."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"* While Nova today tracks the capabilities of network interfaces in the"},{"line_number":52,"context_line":"  ``extra_info`` field of the ``pci_devices`` table and the"}],"source_content_type":"text/x-rst","patch_set":3,"id":"e34851ed_6a8bcca1","line":49,"range":{"start_line":49,"start_character":31,"end_line":49,"end_character":50},"updated":"2023-06-08 09:57:01.000000000","message":"are you sure ? in general, the scheduler can be very quick, even by doing python filtering and the recent history showed that the benefits of Placement queries were absolute a feat but not necessarly in terms of rapidity.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"372f5b6fcfea37784426f2ea1263d41edbc72639","unresolved":true,"context_lines":[{"line_number":46,"context_line":""},{"line_number":47,"context_line":".. rubric:: Limitations"},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"* The current implementation is functionally slow."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"* While Nova today tracks the capabilities of network interfaces in the"},{"line_number":52,"context_line":"  ``extra_info`` field of the ``pci_devices`` table and the"}],"source_content_type":"text/x-rst","patch_set":3,"id":"e6a7fb1b_64534eeb","line":49,"range":{"start_line":49,"start_character":31,"end_line":49,"end_character":50},"in_reply_to":"e34851ed_6a8bcca1","updated":"2023-06-08 11:17:17.000000000","message":"its not the pci part thats slow its the numa toplogy filter that runs after.\n\nthe pci filter is actully pretty quick.\n\nif the vm also has a numa toplogy and the numa toplogy filter need to run it can be slow becaues it has to choose up to n numa nodes form m host numa nodes consider ing many factors \n\nthe pci filter just checks if there are devices avaiable of the appporate type on the required physnet.\n\nthats all pretty liniar and will be comperably to placement.\n\nplacement should be slightly faster but untill we also track numa locality\n\nso there should be a minor speedup for non numa instances and very little differnce for numa ones.\n\nthe real benift is not speed its correctness.\n\ncern i belive limit the placement respoce to somethign like 10 candiates.\n\nsince pci devices are not currently tracked in placment for nics its entrily possble that none of the 10 hosts will have pci devices aviable\nand they will get a no valid host even when there are host aviable.\n\n\nThis would currenlty requrie you to ask for more candiates form placment\nso the actual speed imporvment of tracking the pci devices in placment will come form two things. first since placement will now consider the pci requests in its query you can reduce the amount of candiates it need to return saving on computation and bandwith. second since you can now have a smaller allocation candiate set the schduler need ot process less host in general.\n\nthat is where the performace imporvement is coming form. its not really because the filter is expensive.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"6a5a48d4e283dab455c5c4f7352f7161cbc3f2c5","unresolved":false,"context_lines":[{"line_number":46,"context_line":""},{"line_number":47,"context_line":".. rubric:: Limitations"},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"* The current implementation is functionally slow."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"* While Nova today tracks the capabilities of network interfaces in the"},{"line_number":52,"context_line":"  ``extra_info`` field of the ``pci_devices`` table and the"}],"source_content_type":"text/x-rst","patch_set":3,"id":"59eb992a_acfeda54","line":49,"range":{"start_line":49,"start_character":31,"end_line":49,"end_character":50},"in_reply_to":"e6a7fb1b_64534eeb","updated":"2023-06-09 05:28:28.000000000","message":"Add short description w.r.t. Numa topology filter","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":true,"context_lines":[{"line_number":53,"context_line":"  ``PciPassthroughFilter`` could match on those capabilities there is no"},{"line_number":54,"context_line":"  user-facing way to express a request for an SR-IOV Neutron port with a"},{"line_number":55,"context_line":"  specific network capability e.g. TSO."},{"line_number":56,"context_line":""},{"line_number":57,"context_line":".. rubric:: Optimizations"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"* Use Placement to track SR-IOV nics."}],"source_content_type":"text/x-rst","patch_set":3,"id":"6ab3c694_e8581a7f","line":56,"updated":"2023-06-08 09:57:01.000000000","message":"honestly, you\u0027re missing the main limitation : we are only claiming very late in the workflow while we could have way less race conditions if we were doing that effort.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"372f5b6fcfea37784426f2ea1263d41edbc72639","unresolved":true,"context_lines":[{"line_number":53,"context_line":"  ``PciPassthroughFilter`` could match on those capabilities there is no"},{"line_number":54,"context_line":"  user-facing way to express a request for an SR-IOV Neutron port with a"},{"line_number":55,"context_line":"  specific network capability e.g. TSO."},{"line_number":56,"context_line":""},{"line_number":57,"context_line":".. rubric:: Optimizations"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"* Use Placement to track SR-IOV nics."}],"source_content_type":"text/x-rst","patch_set":3,"id":"a7f0a781_792d35e3","line":56,"in_reply_to":"6ab3c694_e8581a7f","updated":"2023-06-08 11:17:17.000000000","message":"yep today onless its a live migration we only claim the pci device on the comptue node so two parallel request can race for the last device on a host and cause a retry form the alternnitv hosts list.\n\nthat is the other performace impovement as we will be elimiating that race entirly.\nwhich should mean no more retryes for pci devices which is a masive win.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"6a5a48d4e283dab455c5c4f7352f7161cbc3f2c5","unresolved":false,"context_lines":[{"line_number":53,"context_line":"  ``PciPassthroughFilter`` could match on those capabilities there is no"},{"line_number":54,"context_line":"  user-facing way to express a request for an SR-IOV Neutron port with a"},{"line_number":55,"context_line":"  specific network capability e.g. TSO."},{"line_number":56,"context_line":""},{"line_number":57,"context_line":".. rubric:: Optimizations"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"* Use Placement to track SR-IOV nics."}],"source_content_type":"text/x-rst","patch_set":3,"id":"340b11d6_c5cc2014","line":56,"in_reply_to":"a7f0a781_792d35e3","updated":"2023-06-09 05:28:28.000000000","message":"Done","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":true,"context_lines":[{"line_number":71,"context_line":""},{"line_number":72,"context_line":".. note::"},{"line_number":73,"context_line":""},{"line_number":74,"context_line":"  Device quotas would require unified limits to be implemented. Implementing"},{"line_number":75,"context_line":"  quotas is out of the scope of this spec beyond enabling the use case by"},{"line_number":76,"context_line":"  modeling PCI devices in Placement."},{"line_number":77,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"7f56fbe6_87ecbd5b","line":74,"range":{"start_line":74,"start_character":22,"end_line":74,"end_character":62},"updated":"2023-06-08 09:57:01.000000000","message":"good news, this is a done work, but we haven\u0027t really provided yet a migration tool for helping operators to transition from the legacy quota model to the new one.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"6a5a48d4e283dab455c5c4f7352f7161cbc3f2c5","unresolved":false,"context_lines":[{"line_number":71,"context_line":""},{"line_number":72,"context_line":".. note::"},{"line_number":73,"context_line":""},{"line_number":74,"context_line":"  Device quotas would require unified limits to be implemented. Implementing"},{"line_number":75,"context_line":"  quotas is out of the scope of this spec beyond enabling the use case by"},{"line_number":76,"context_line":"  modeling PCI devices in Placement."},{"line_number":77,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"93a7c86c_18abe2f2","line":74,"range":{"start_line":74,"start_character":22,"end_line":74,"end_character":62},"in_reply_to":"7f56fbe6_87ecbd5b","updated":"2023-06-09 05:28:28.000000000","message":"Done","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":false,"context_lines":[{"line_number":102,"context_line":"to model these resources. The resource classes are explained in later section."},{"line_number":103,"context_line":""},{"line_number":104,"context_line":"Today Nova allows consuming type-PCI or type-VF for ``direct`` ports. This is"},{"line_number":105,"context_line":"mostly there due to historical reasons and it should be cleaned up."},{"line_number":106,"context_line":""},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"Modeling SR-IOV devices in Placement"}],"source_content_type":"text/x-rst","patch_set":3,"id":"e02273ae_a6245a6e","line":105,"updated":"2023-06-08 09:57:01.000000000","message":"++","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"372f5b6fcfea37784426f2ea1263d41edbc72639","unresolved":false,"context_lines":[{"line_number":102,"context_line":"to model these resources. The resource classes are explained in later section."},{"line_number":103,"context_line":""},{"line_number":104,"context_line":"Today Nova allows consuming type-PCI or type-VF for ``direct`` ports. This is"},{"line_number":105,"context_line":"mostly there due to historical reasons and it should be cleaned up."},{"line_number":106,"context_line":""},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"Modeling SR-IOV devices in Placement"}],"source_content_type":"text/x-rst","patch_set":3,"id":"de436a28_daf62ffb","line":105,"in_reply_to":"e02273ae_a6245a6e","updated":"2023-06-08 11:17:17.000000000","message":"to be clear we will still need to keep the disticntion in the nova db.\n\nbut neutron ports willl only consome type-vf going forward since using type-pci for a direct prot is currently broken anyway.\n\ntype-vf and type-pci is still imporant for flavor alais based pci passthough so this is just a statement about how we will model them in placment. the nova part will remain the same.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":false,"context_lines":[{"line_number":113,"context_line":"provider (RP) with the name ``\u003chypervisor_hostname\u003e_\u003cpci_address\u003e``. The"},{"line_number":114,"context_line":"hypervisor_hostname prefix will be the same string as the name of the root"},{"line_number":115,"context_line":"RP. The pci_address part of the name will be the full PCI address in the"},{"line_number":116,"context_line":"same format of ``DDDD:BB:AA.FF``."},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"Each SR-IOV NIC device RP will have an inventory of resource class and traits"},{"line_number":119,"context_line":"derived by Nova based on device categorization explained below:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3327ffe0_09d2a9e3","line":116,"range":{"start_line":116,"start_character":16,"end_line":116,"end_character":32},"updated":"2023-06-08 09:57:01.000000000","message":"can\u0027t remember if the pci address is somehow translated using underscores, but meh.","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"372f5b6fcfea37784426f2ea1263d41edbc72639","unresolved":false,"context_lines":[{"line_number":113,"context_line":"provider (RP) with the name ``\u003chypervisor_hostname\u003e_\u003cpci_address\u003e``. The"},{"line_number":114,"context_line":"hypervisor_hostname prefix will be the same string as the name of the root"},{"line_number":115,"context_line":"RP. The pci_address part of the name will be the full PCI address in the"},{"line_number":116,"context_line":"same format of ``DDDD:BB:AA.FF``."},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"Each SR-IOV NIC device RP will have an inventory of resource class and traits"},{"line_number":119,"context_line":"derived by Nova based on device categorization explained below:"}],"source_content_type":"text/x-rst","patch_set":3,"id":"e9e84d46_51296fea","line":116,"range":{"start_line":116,"start_character":16,"end_line":116,"end_character":32},"in_reply_to":"3327ffe0_09d2a9e3","updated":"2023-06-08 11:17:17.000000000","message":"i dont think it is today but we will be keeping the same format as we use for non nics in placment","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"a259b92c485a695edfff5f83407024de57620192","unresolved":false,"context_lines":[{"line_number":176,"context_line":"  409 until support for this is added."},{"line_number":177,"context_line":""},{"line_number":178,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":179,"context_line":"  rejected."},{"line_number":180,"context_line":""},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"bf830cbc_a4ef85a5","line":179,"updated":"2023-06-08 09:57:01.000000000","message":"++","commit_id":"afee2f95c91a9dbff3823637ba6c37754cb6bdd9"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"b4c02984fdf574175de7877a935a92eb1b876949","unresolved":true,"context_lines":[{"line_number":177,"context_line":"Please note, Nova will consider below mentioned scenarios only if"},{"line_number":178,"context_line":"``[filter_scheduler]pci_in_placement`` config option is enabled:"},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"* Nova will detect and refuse to boot an instance with an SR-IOV type port that"},{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."}],"source_content_type":"text/x-rst","patch_set":4,"id":"33c0ff58_ca3b2a94","line":182,"range":{"start_line":180,"start_character":2,"end_line":182,"end_character":38},"updated":"2023-06-13 23:36:37.000000000","message":"Can we detect that early enough to reject it synchronously? We read the ports while we are still sync, but nova-api does not have [filter_scheduler]pci_in_placement config, I think that only used by / needed by the scheduler. But the scheduler is already called async.","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"ab4a4be32ba8e5696405bbbe0417e01ba15714f3","unresolved":false,"context_lines":[{"line_number":177,"context_line":"Please note, Nova will consider below mentioned scenarios only if"},{"line_number":178,"context_line":"``[filter_scheduler]pci_in_placement`` config option is enabled:"},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"* Nova will detect and refuse to boot an instance with an SR-IOV type port that"},{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."}],"source_content_type":"text/x-rst","patch_set":4,"id":"e1387dfc_26853236","line":182,"range":{"start_line":180,"start_character":2,"end_line":182,"end_character":38},"in_reply_to":"2219c446_67672dd8","updated":"2023-06-30 12:02:02.000000000","message":"So you mean it will be a NoValidHost type of error?","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"665db372cc63779ca6a338501bb383791ad7ef9a","unresolved":false,"context_lines":[{"line_number":177,"context_line":"Please note, Nova will consider below mentioned scenarios only if"},{"line_number":178,"context_line":"``[filter_scheduler]pci_in_placement`` config option is enabled:"},{"line_number":179,"context_line":""},{"line_number":180,"context_line":"* Nova will detect and refuse to boot an instance with an SR-IOV type port that"},{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."}],"source_content_type":"text/x-rst","patch_set":4,"id":"2219c446_67672dd8","line":182,"range":{"start_line":180,"start_character":2,"end_line":182,"end_character":38},"in_reply_to":"33c0ff58_ca3b2a94","updated":"2023-06-14 08:25:25.000000000","message":"Done","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"b4c02984fdf574175de7877a935a92eb1b876949","unresolved":true,"context_lines":[{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":""},{"line_number":188,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"}],"source_content_type":"text/x-rst","patch_set":4,"id":"e4db9e55_de99c4bd","line":185,"range":{"start_line":184,"start_character":2,"end_line":185,"end_character":11},"updated":"2023-06-13 23:36:37.000000000","message":"That does not go through the scheduler today (?!) and therefore you cannot use [filter_scheduler]pci_in_placement to decide if it is allowed","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"ab4a4be32ba8e5696405bbbe0417e01ba15714f3","unresolved":true,"context_lines":[{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":""},{"line_number":188,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bf3e74f2_877ae980","line":185,"range":{"start_line":184,"start_character":2,"end_line":185,"end_character":11},"in_reply_to":"befef1a4_b9773860","updated":"2023-06-30 12:02:02.000000000","message":"That is fairly non trivial. So you need to expand this in the spec what are the implication of that change.","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"14f3093e5ca1124aa3c02b97a4de66b4ee5dbc3c","unresolved":true,"context_lines":[{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":""},{"line_number":188,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"}],"source_content_type":"text/x-rst","patch_set":4,"id":"33a63c82_f8f582b8","line":185,"range":{"start_line":184,"start_character":2,"end_line":185,"end_character":11},"in_reply_to":"bf3e74f2_877ae980","updated":"2023-07-04 16:16:48.000000000","message":"\u003e 14:43 \u003ckpdev\u003e wha you think interface attach call with sriov ports + QoS will be \u003e non trivial.\n\u003e 14:43 \u003ckpdev\u003e what issues you have in your mind ?\n\nIf you need to call the scheduler just to reject the interface attach with QoS when `[filter_scheduler]pci_in_placement` is true, then you have to change how interface attach is called from, as today it is not calling the scheduler. This means an RPC change that needs to be detailed out.","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"665db372cc63779ca6a338501bb383791ad7ef9a","unresolved":true,"context_lines":[{"line_number":181,"context_line":"  contains addtional resource requests. This will be done by returning a code"},{"line_number":182,"context_line":"  409 until support for this is added."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"* Attaching SR-IOV ports with additional resources will also be detected and"},{"line_number":185,"context_line":"  rejected."},{"line_number":186,"context_line":""},{"line_number":187,"context_line":""},{"line_number":188,"context_line":"Neutron SR-IOV ports with QoS (out of scope)"}],"source_content_type":"text/x-rst","patch_set":4,"id":"befef1a4_b9773860","line":185,"range":{"start_line":184,"start_character":2,"end_line":185,"end_character":11},"in_reply_to":"e4db9e55_de99c4bd","updated":"2023-06-14 08:25:25.000000000","message":"we might need to route through scheduler then.","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"b4c02984fdf574175de7877a935a92eb1b876949","unresolved":true,"context_lines":[{"line_number":214,"context_line":"  RP tree. By that, we could make the bandwidth RP a sharing RP that shares"},{"line_number":215,"context_line":"  resources with the PCI device RP representing the physical device."},{"line_number":216,"context_line":""},{"line_number":217,"context_line":"To enable forward progress with the minimum of dependencies and incremental"},{"line_number":218,"context_line":"progress, the preferred short term solution is to enhance the existing"},{"line_number":219,"context_line":"pci_passthough_filter to remove allocation candidates where these request"},{"line_number":220,"context_line":"groups are fulfilled from different physical devices or add a new scheduler"},{"line_number":221,"context_line":"filter that removes allocation candidates."},{"line_number":222,"context_line":""},{"line_number":223,"context_line":""},{"line_number":224,"context_line":"Requesting PCI devices"}],"source_content_type":"text/x-rst","patch_set":4,"id":"90aa439a_af342ab1","line":221,"range":{"start_line":217,"start_character":0,"end_line":221,"end_character":42},"updated":"2023-06-13 23:36:37.000000000","message":"If we reject the requests or declare such requests as not supported then we don\u0027t need to implement this extra filtering. If we implement this extra filtering then we actually implement the feature","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"665db372cc63779ca6a338501bb383791ad7ef9a","unresolved":false,"context_lines":[{"line_number":214,"context_line":"  RP tree. By that, we could make the bandwidth RP a sharing RP that shares"},{"line_number":215,"context_line":"  resources with the PCI device RP representing the physical device."},{"line_number":216,"context_line":""},{"line_number":217,"context_line":"To enable forward progress with the minimum of dependencies and incremental"},{"line_number":218,"context_line":"progress, the preferred short term solution is to enhance the existing"},{"line_number":219,"context_line":"pci_passthough_filter to remove allocation candidates where these request"},{"line_number":220,"context_line":"groups are fulfilled from different physical devices or add a new scheduler"},{"line_number":221,"context_line":"filter that removes allocation candidates."},{"line_number":222,"context_line":""},{"line_number":223,"context_line":""},{"line_number":224,"context_line":"Requesting PCI devices"}],"source_content_type":"text/x-rst","patch_set":4,"id":"0271e689_abd75e69","line":221,"range":{"start_line":217,"start_character":0,"end_line":221,"end_character":42},"in_reply_to":"90aa439a_af342ab1","updated":"2023-06-14 08:25:25.000000000","message":"Done","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"b4c02984fdf574175de7877a935a92eb1b876949","unresolved":true,"context_lines":[{"line_number":297,"context_line":"the result of the offloading of the filtering to Placement and the removal of"},{"line_number":298,"context_line":"reschedules due to racing for the last PCI device on a host, the overall"},{"line_number":299,"context_line":"performance is expected to improve."},{"line_number":300,"context_line":""},{"line_number":301,"context_line":"Other deployer impact"},{"line_number":302,"context_line":"---------------------"},{"line_number":303,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"a0719cb7_7214d1f5","line":300,"updated":"2023-06-13 23:36:37.000000000","message":"You need to mention that there is a performance issue generating many allocation candidates. https://review.opendev.org/c/openstack/nova/+/855885","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":35507,"name":"Kiran Pawar","email":"kiranpawarpict2010@gmail.com","username":"kpawar"},"change_message_id":"665db372cc63779ca6a338501bb383791ad7ef9a","unresolved":false,"context_lines":[{"line_number":297,"context_line":"the result of the offloading of the filtering to Placement and the removal of"},{"line_number":298,"context_line":"reschedules due to racing for the last PCI device on a host, the overall"},{"line_number":299,"context_line":"performance is expected to improve."},{"line_number":300,"context_line":""},{"line_number":301,"context_line":"Other deployer impact"},{"line_number":302,"context_line":"---------------------"},{"line_number":303,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"0527feb1_9c53e2ea","line":300,"in_reply_to":"a0719cb7_7214d1f5","updated":"2023-06-14 08:25:25.000000000","message":"Done","commit_id":"96ad7c927dfe031169962236439242a73ab80a5c"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"ab4a4be32ba8e5696405bbbe0417e01ba15714f3","unresolved":true,"context_lines":[{"line_number":302,"context_line":""},{"line_number":303,"context_line":"In case of multiple devices are requested, there is possibility of many"},{"line_number":304,"context_line":"allocation candidates will be generated by placement. This will considerably"},{"line_number":305,"context_line":"slow down the PCI device selection workflow."},{"line_number":306,"context_line":""},{"line_number":307,"context_line":"Other deployer impact"},{"line_number":308,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":5,"id":"446e3a84_8a73cf1f","line":305,"updated":"2023-06-30 12:02:02.000000000","message":"Please link to the patch showing the magnitude of the issue. https://review.opendev.org/c/openstack/nova/+/855885","commit_id":"5f3d549cfbc23b7490a94c326d6ea29cebfd867c"}]}