)]}'
{"specs/wallaby/approved/modernize-os-hypervisors-api.rst":[{"author":{"_account_id":10135,"name":"Lee Yarwood","display_name":"Lee Yarwood","email":"lyarwood@redhat.com","username":"lyarwood"},"change_message_id":"b027fdfacebdefda31597d4b42cb2aea1045e9d2","unresolved":false,"context_lines":[{"line_number":5,"context_line":" http://creativecommons.org/licenses/by/3.0/legalcode"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":8,"context_line":"Remove resource information from ``os-hypervisors`` API"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/modernize-os-hypervisors-api"}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_5ce20d04","line":8,"range":{"start_line":8,"start_character":33,"end_line":8,"end_character":51},"updated":"2020-09-30 09:34:18.000000000","message":"https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_ead/755109/2/check/openstack-tox-docs/ead5d87/docs/specs/wallaby/index.html\n\nThis does result in os-hypervisors being listed in bold in the spec index page, not that I care it just looks odd.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"1b768f33b301b9a5d61dd446f54d03d8ccfd7a8b","unresolved":false,"context_lines":[{"line_number":98,"context_line":""},{"line_number":99,"context_line":"- ``cpu_info``"},{"line_number":100,"context_line":""},{"line_number":101,"context_line":"  Useful at face value but irrelevant in practice outside of requests for"},{"line_number":102,"context_line":"  specific CPU features, which are already handled by placement trait requests."},{"line_number":103,"context_line":""},{"line_number":104,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``, ``disk_available_least``"},{"line_number":105,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_5cffad34","line":102,"range":{"start_line":101,"start_character":0,"end_line":102,"end_character":79},"updated":"2020-09-30 10:18:00.000000000","message":"As far as I know the topology part is not covered by placement.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"7d2d25b75c1a2a349cb00617e63d54200a794071","unresolved":false,"context_lines":[{"line_number":98,"context_line":""},{"line_number":99,"context_line":"- ``cpu_info``"},{"line_number":100,"context_line":""},{"line_number":101,"context_line":"  Useful at face value but irrelevant in practice outside of requests for"},{"line_number":102,"context_line":"  specific CPU features, which are already handled by placement trait requests."},{"line_number":103,"context_line":""},{"line_number":104,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``, ``disk_available_least``"},{"line_number":105,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_1f7235b1","line":102,"range":{"start_line":101,"start_character":0,"end_line":102,"end_character":79},"in_reply_to":"9f560f44_5cffad34","updated":"2020-10-01 11:13:34.000000000","message":"True, but it\u0027s not relevant to scheduling since you can\u0027t use it. Updated the text here.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":10135,"name":"Lee Yarwood","display_name":"Lee Yarwood","email":"lyarwood@redhat.com","username":"lyarwood"},"change_message_id":"b027fdfacebdefda31597d4b42cb2aea1045e9d2","unresolved":false,"context_lines":[{"line_number":103,"context_line":""},{"line_number":104,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``, ``disk_available_least``"},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":107,"context_line":"  overcommit into account. Use placement."},{"line_number":108,"context_line":""},{"line_number":109,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_7c1431ea","line":106,"range":{"start_line":106,"start_character":2,"end_line":106,"end_character":4},"updated":"2020-09-30 09:34:18.000000000","message":"\\o/","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"1b768f33b301b9a5d61dd446f54d03d8ccfd7a8b","unresolved":false,"context_lines":[{"line_number":104,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``, ``disk_available_least``"},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":107,"context_line":"  overcommit into account. Use placement."},{"line_number":108,"context_line":""},{"line_number":109,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":110,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_fc14c14f","line":107,"range":{"start_line":107,"start_character":27,"end_line":107,"end_character":41},"updated":"2020-09-30 10:18:00.000000000","message":"disk_available_least does not have a placement based replacement as it is not about the disk inventory - disk allocation, it is actual disk usage on the hypervisor. Including the case when the host OS uses more space that what is reserved for it. Including the case when an instance allocates in placement 10G but only use 1G actually on the host. Placement simply not created to hold actual usage information it holds only inventory and allocation.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"7d2d25b75c1a2a349cb00617e63d54200a794071","unresolved":false,"context_lines":[{"line_number":104,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``, ``disk_available_least``"},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":107,"context_line":"  overcommit into account. Use placement."},{"line_number":108,"context_line":""},{"line_number":109,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":110,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_bfdfc964","line":107,"range":{"start_line":107,"start_character":27,"end_line":107,"end_character":41},"in_reply_to":"9f560f44_fc14c14f","updated":"2020-10-01 11:13:34.000000000","message":"Okay, good point. I still don\u0027t think this is actionable and should be removed, but I\u0027ve split this out into its own field with an explanation.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":10135,"name":"Lee Yarwood","display_name":"Lee Yarwood","email":"lyarwood@redhat.com","username":"lyarwood"},"change_message_id":"b027fdfacebdefda31597d4b42cb2aea1045e9d2","unresolved":false,"context_lines":[{"line_number":143,"context_line":"REST API impact"},{"line_number":144,"context_line":"---------------"},{"line_number":145,"context_line":""},{"line_number":146,"context_line":"The ``/os-hypervisors/detail`` API will no longer include the following fields"},{"line_number":147,"context_line":"in it\u0027s response: ``cpu_info``, ``free_disk_gb``, ``local_gb``,"},{"line_number":148,"context_line":"``local_gb_used``, ``disk_available_least``, ``free_ram_mb``, ``memory_mb``,"},{"line_number":149,"context_line":"``memory_mb_used`` and ``running_vms``."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"The ``/os-hypervisors/statistics`` API will be removed entirely."},{"line_number":152,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_5cf4cdd0","line":149,"range":{"start_line":146,"start_character":0,"end_line":149,"end_character":39},"updated":"2020-09-30 09:34:18.000000000","message":"nit - I\u0027d reference that these are only dropped from a new microversion.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"7d2d25b75c1a2a349cb00617e63d54200a794071","unresolved":false,"context_lines":[{"line_number":143,"context_line":"REST API impact"},{"line_number":144,"context_line":"---------------"},{"line_number":145,"context_line":""},{"line_number":146,"context_line":"The ``/os-hypervisors/detail`` API will no longer include the following fields"},{"line_number":147,"context_line":"in it\u0027s response: ``cpu_info``, ``free_disk_gb``, ``local_gb``,"},{"line_number":148,"context_line":"``local_gb_used``, ``disk_available_least``, ``free_ram_mb``, ``memory_mb``,"},{"line_number":149,"context_line":"``memory_mb_used`` and ``running_vms``."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"The ``/os-hypervisors/statistics`` API will be removed entirely."},{"line_number":152,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_dfbb9d0c","line":149,"range":{"start_line":146,"start_character":0,"end_line":149,"end_character":39},"in_reply_to":"9f560f44_5cf4cdd0","updated":"2020-10-01 11:13:34.000000000","message":"Done","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":10135,"name":"Lee Yarwood","display_name":"Lee Yarwood","email":"lyarwood@redhat.com","username":"lyarwood"},"change_message_id":"b027fdfacebdefda31597d4b42cb2aea1045e9d2","unresolved":false,"context_lines":[{"line_number":148,"context_line":"``local_gb_used``, ``disk_available_least``, ``free_ram_mb``, ``memory_mb``,"},{"line_number":149,"context_line":"``memory_mb_used`` and ``running_vms``."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"The ``/os-hypervisors/statistics`` API will be removed entirely."},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Security impact"},{"line_number":154,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_bc0529c4","line":151,"range":{"start_line":151,"start_character":0,"end_line":151,"end_character":64},"updated":"2020-09-30 09:34:18.000000000","message":"nit - removed by returning 410 after a newly introduced microversion right?","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"7d2d25b75c1a2a349cb00617e63d54200a794071","unresolved":false,"context_lines":[{"line_number":148,"context_line":"``local_gb_used``, ``disk_available_least``, ``free_ram_mb``, ``memory_mb``,"},{"line_number":149,"context_line":"``memory_mb_used`` and ``running_vms``."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"The ``/os-hypervisors/statistics`` API will be removed entirely."},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Security impact"},{"line_number":154,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_bfb86911","line":151,"range":{"start_line":151,"start_character":0,"end_line":151,"end_character":64},"in_reply_to":"9f560f44_bc0529c4","updated":"2020-10-01 11:13:34.000000000","message":"Done","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"1b768f33b301b9a5d61dd446f54d03d8ccfd7a8b","unresolved":false,"context_lines":[{"line_number":164,"context_line":"---------------------"},{"line_number":165,"context_line":""},{"line_number":166,"context_line":"The clients will need to be updated. Documentation referencing these APIs will"},{"line_number":167,"context_line":"need to be updated with recommendations to look at placement instead."},{"line_number":168,"context_line":""},{"line_number":169,"context_line":"Performance Impact"},{"line_number":170,"context_line":"------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_dcb9fd7c","line":167,"updated":"2020-09-30 10:18:00.000000000","message":"We might even want to give specific examples how to get the  information from placement that is not r eturned by the os-hypervisors any more.","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"7d2d25b75c1a2a349cb00617e63d54200a794071","unresolved":false,"context_lines":[{"line_number":164,"context_line":"---------------------"},{"line_number":165,"context_line":""},{"line_number":166,"context_line":"The clients will need to be updated. Documentation referencing these APIs will"},{"line_number":167,"context_line":"need to be updated with recommendations to look at placement instead."},{"line_number":168,"context_line":""},{"line_number":169,"context_line":"Performance Impact"},{"line_number":170,"context_line":"------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"9f560f44_7fdff137","line":167,"in_reply_to":"9f560f44_dcb9fd7c","updated":"2020-10-01 11:13:34.000000000","message":"Done","commit_id":"197b178ee43f8ba0269cb744e1e98f453bc01173"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":5,"context_line":" http://creativecommons.org/licenses/by/3.0/legalcode"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":8,"context_line":"Remove resource information from ``os-hypervisors`` API"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/modernize-os-hypervisors-api"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_2e2ed1f9","line":8,"range":{"start_line":8,"start_character":0,"end_line":8,"end_character":55},"updated":"2020-10-27 17:59:12.000000000","message":"we have a ptg topic regardign deprecating this entirely\nand removing after W so i guess this prosposal is the altrenitive in the event we do not depreccate the api?","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"1d908fff86000dea77fb7cc291847eb7dd37d43e","unresolved":false,"context_lines":[{"line_number":5,"context_line":" http://creativecommons.org/licenses/by/3.0/legalcode"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":8,"context_line":"Remove resource information from ``os-hypervisors`` API"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/modernize-os-hypervisors-api"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_f1df98d3","line":8,"range":{"start_line":8,"start_character":0,"end_line":8,"end_character":55},"in_reply_to":"3f65232a_2e2ed1f9","updated":"2020-10-27 18:27:13.000000000","message":"This spec initially started life as \"Remove ``os-hypervisors`` API\", but it evolved as I discovered things that still made sense to report. See below for the breakdown","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":16,"context_line":"it\u0027s a shell of itself. The more detailed information it returns is"},{"line_number":17,"context_line":"hypervisor-specific and frequently wrong, especially with advanced features"},{"line_number":18,"context_line":"like CPU pinning or file-based memory. With the elevation of placement to its"},{"line_number":19,"context_line":"rightful place as lord of (almost) all things resource\u0027y, the ``os-hypervisor``"},{"line_number":20,"context_line":"API needs to slim down significantly."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"Problem description"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_6fbdaa67","line":19,"range":{"start_line":19,"start_character":27,"end_line":19,"end_character":33},"updated":"2020-10-21 16:23:13.000000000","message":"heh, you\u0027re fortunate as the resources you mentioned in the previous sentence are now modeled in placement.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":79,"context_line":""},{"line_number":80,"context_line":"The fields here broadly fall into three categories: useful but duplicated in"},{"line_number":81,"context_line":"the summary (non-detailed) view, useful and unique to the detailed view, and"},{"line_number":82,"context_line":"not useful. First, the useful but duplicated fields:"},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"- ``id``"},{"line_number":85,"context_line":"- ``status``"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_4f670e27","line":82,"range":{"start_line":82,"start_character":12,"end_line":82,"end_character":52},"updated":"2020-10-21 16:23:13.000000000","message":"I don\u0027t really see the need of describing the fields that are also present on the summary view, as we do it also for other API resources.\nI just feel confused whether you want to change this or not.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"8ec36c4ec8a11fd1d6c02f308d50cb06deb945ec","unresolved":false,"context_lines":[{"line_number":79,"context_line":""},{"line_number":80,"context_line":"The fields here broadly fall into three categories: useful but duplicated in"},{"line_number":81,"context_line":"the summary (non-detailed) view, useful and unique to the detailed view, and"},{"line_number":82,"context_line":"not useful. First, the useful but duplicated fields:"},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"- ``id``"},{"line_number":85,"context_line":"- ``status``"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_e5f719d4","line":82,"range":{"start_line":82,"start_character":12,"end_line":82,"end_character":52},"in_reply_to":"3f65232a_4f670e27","updated":"2020-10-21 17:37:30.000000000","message":"Ah, I wasn\u0027t clear. What I meant to say is that I\u0027ve no issues with the summary view so we can ignore those fields.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"b140fb3b680e88ffc25c46c35661b6b573aedaf7","unresolved":false,"context_lines":[{"line_number":79,"context_line":""},{"line_number":80,"context_line":"The fields here broadly fall into three categories: useful but duplicated in"},{"line_number":81,"context_line":"the summary (non-detailed) view, useful and unique to the detailed view, and"},{"line_number":82,"context_line":"not useful. First, the useful but duplicated fields:"},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"- ``id``"},{"line_number":85,"context_line":"- ``status``"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_a05161db","line":82,"range":{"start_line":82,"start_character":12,"end_line":82,"end_character":52},"in_reply_to":"3f65232a_e5f719d4","updated":"2020-10-26 11:09:10.000000000","message":"Oh ok, all good then.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":92,"context_line":"- ``hypervisor_type``"},{"line_number":93,"context_line":"- ``hypervisor_version``"},{"line_number":94,"context_line":"- ``service``"},{"line_number":95,"context_line":"- ``current_workload``"},{"line_number":96,"context_line":""},{"line_number":97,"context_line":"Finally, the useless fields:"},{"line_number":98,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_ae678151","line":95,"range":{"start_line":95,"start_character":4,"end_line":95,"end_character":20},"updated":"2020-10-27 17:59:12.000000000","message":"is this useful. i assume it the current vm count?\nyou could get that via the server list api with --host.\nit perhaps semi useful but i am not sure why it would not be in the summary.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"c1da65ee7f9140b77da7aa5d17dd48eec95afcc8","unresolved":false,"context_lines":[{"line_number":92,"context_line":"- ``hypervisor_type``"},{"line_number":93,"context_line":"- ``hypervisor_version``"},{"line_number":94,"context_line":"- ``service``"},{"line_number":95,"context_line":"- ``current_workload``"},{"line_number":96,"context_line":""},{"line_number":97,"context_line":"Finally, the useless fields:"},{"line_number":98,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"1f621f24_765bd8f4","line":95,"range":{"start_line":95,"start_character":4,"end_line":95,"end_character":20},"in_reply_to":"3f65232a_91eaa472","updated":"2020-11-17 12:35:59.000000000","message":"Done","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"1d908fff86000dea77fb7cc291847eb7dd37d43e","unresolved":false,"context_lines":[{"line_number":92,"context_line":"- ``hypervisor_type``"},{"line_number":93,"context_line":"- ``hypervisor_version``"},{"line_number":94,"context_line":"- ``service``"},{"line_number":95,"context_line":"- ``current_workload``"},{"line_number":96,"context_line":""},{"line_number":97,"context_line":"Finally, the useless fields:"},{"line_number":98,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_91eaa472","line":95,"range":{"start_line":95,"start_character":4,"end_line":95,"end_character":20},"in_reply_to":"3f65232a_ae678151","updated":"2020-10-27 18:27:13.000000000","message":"This is \"the number of tasks the hypervisor is responsible for\" and it \"will be equal or greater than the number of active VMs on the system (it can be greater when VMs are being deleted and the hypervisor is still cleaning up)\".\n\nWhen I wrote this, I actually though this was more nuanced and covered things like IO workloads. Assuming not, then you\u0027re correct and this could be replaced with \u0027server list --host\u0027 also and can be removed from here.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":101,"context_line":"  Useful at face value but the only thing relevant for scheduling purposes are"},{"line_number":102,"context_line":"  the CPU architecture and CPU features, all of which are already handled by"},{"line_number":103,"context_line":"  placement trait requests. The topology field is an oddity that should likely"},{"line_number":104,"context_line":"  never have been added. It\u0027s not usable in scheduling and is possibly wrong,"},{"line_number":105,"context_line":"  given it doesn\u0027t reflect offline CPUs or those not available to nova due to"},{"line_number":106,"context_line":"  configuration, and it doesn\u0027t handle non-uniform CPU topologies where there"},{"line_number":107,"context_line":"  are e.g. more cores on one socket than another. If the operator wants this"},{"line_number":108,"context_line":"  information, they can simply inspect the host like they would have to do to"},{"line_number":109,"context_line":"  identify e.g. the specifics of PCI devices or storage devices."},{"line_number":110,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_2f59f259","line":107,"range":{"start_line":104,"start_character":25,"end_line":107,"end_character":49},"updated":"2020-10-21 16:23:13.000000000","message":"While one way I do understand your point, on the other way, I don\u0027t see the justification to remove such information until we get it modeled more accurately in Placement.\n\nEven if this topology doesn\u0027t reflect the available NUMA resources for the host, ops could use it for fetching information they care directly from an API.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"8ec36c4ec8a11fd1d6c02f308d50cb06deb945ec","unresolved":false,"context_lines":[{"line_number":101,"context_line":"  Useful at face value but the only thing relevant for scheduling purposes are"},{"line_number":102,"context_line":"  the CPU architecture and CPU features, all of which are already handled by"},{"line_number":103,"context_line":"  placement trait requests. The topology field is an oddity that should likely"},{"line_number":104,"context_line":"  never have been added. It\u0027s not usable in scheduling and is possibly wrong,"},{"line_number":105,"context_line":"  given it doesn\u0027t reflect offline CPUs or those not available to nova due to"},{"line_number":106,"context_line":"  configuration, and it doesn\u0027t handle non-uniform CPU topologies where there"},{"line_number":107,"context_line":"  are e.g. more cores on one socket than another. If the operator wants this"},{"line_number":108,"context_line":"  information, they can simply inspect the host like they would have to do to"},{"line_number":109,"context_line":"  identify e.g. the specifics of PCI devices or storage devices."},{"line_number":110,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_a5a4010a","line":107,"range":{"start_line":104,"start_character":25,"end_line":107,"end_character":49},"in_reply_to":"3f65232a_2f59f259","updated":"2020-10-21 17:37:30.000000000","message":"What use is it though? It\u0027s not relevant for anything in nova and while it may be useful for ops, we don\u0027t report other things like kernel version, whether the host is up-to-date, what packages are installed etc.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"b140fb3b680e88ffc25c46c35661b6b573aedaf7","unresolved":false,"context_lines":[{"line_number":101,"context_line":"  Useful at face value but the only thing relevant for scheduling purposes are"},{"line_number":102,"context_line":"  the CPU architecture and CPU features, all of which are already handled by"},{"line_number":103,"context_line":"  placement trait requests. The topology field is an oddity that should likely"},{"line_number":104,"context_line":"  never have been added. It\u0027s not usable in scheduling and is possibly wrong,"},{"line_number":105,"context_line":"  given it doesn\u0027t reflect offline CPUs or those not available to nova due to"},{"line_number":106,"context_line":"  configuration, and it doesn\u0027t handle non-uniform CPU topologies where there"},{"line_number":107,"context_line":"  are e.g. more cores on one socket than another. If the operator wants this"},{"line_number":108,"context_line":"  information, they can simply inspect the host like they would have to do to"},{"line_number":109,"context_line":"  identify e.g. the specifics of PCI devices or storage devices."},{"line_number":110,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_405a0df3","line":107,"range":{"start_line":104,"start_character":25,"end_line":107,"end_character":49},"in_reply_to":"3f65232a_a5a4010a","updated":"2020-10-26 11:09:10.000000000","message":"Ok, let\u0027s discuss it at the PTG then.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":108,"context_line":"  information, they can simply inspect the host like they would have to do to"},{"line_number":109,"context_line":"  identify e.g. the specifics of PCI devices or storage devices."},{"line_number":110,"context_line":""},{"line_number":111,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``"},{"line_number":112,"context_line":""},{"line_number":113,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":114,"context_line":"  overcommit into account. Use placement."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"- ``disk_available_least``"},{"line_number":117,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_ef237abd","line":114,"range":{"start_line":111,"start_character":0,"end_line":114,"end_character":41},"updated":"2020-10-21 16:23:13.000000000","message":"I totally agree, those fields should die in fire now that we have a better way to get the correct information.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":113,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":114,"context_line":"  overcommit into account. Use placement."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"- ``disk_available_least``"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"  Reflects the actual disk usage on the hypervisor.  This can help identify"},{"line_number":119,"context_line":"  cases such as a host OS using more space that what is reserved for it or an"},{"line_number":120,"context_line":"  instance not consuming all of the resources allocated to it in placement."},{"line_number":121,"context_line":"  However, these cases either reflect a simple misconfiguration - a need to"},{"line_number":122,"context_line":"  increase the reserved disk space or to potentially enable disk overcommit -"},{"line_number":123,"context_line":"  or are simply useless since they have no impact on scheduling, which only"},{"line_number":124,"context_line":"  care about inventory in placement."},{"line_number":125,"context_line":""},{"line_number":126,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":127,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_cf095e32","line":124,"range":{"start_line":116,"start_character":0,"end_line":124,"end_character":36},"updated":"2020-10-21 16:23:13.000000000","message":"disk_available_least is errorprone. Kill it too.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"c1da65ee7f9140b77da7aa5d17dd48eec95afcc8","unresolved":false,"context_lines":[{"line_number":113,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":114,"context_line":"  overcommit into account. Use placement."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"- ``disk_available_least``"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"  Reflects the actual disk usage on the hypervisor.  This can help identify"},{"line_number":119,"context_line":"  cases such as a host OS using more space that what is reserved for it or an"},{"line_number":120,"context_line":"  instance not consuming all of the resources allocated to it in placement."},{"line_number":121,"context_line":"  However, these cases either reflect a simple misconfiguration - a need to"},{"line_number":122,"context_line":"  increase the reserved disk space or to potentially enable disk overcommit -"},{"line_number":123,"context_line":"  or are simply useless since they have no impact on scheduling, which only"},{"line_number":124,"context_line":"  care about inventory in placement."},{"line_number":125,"context_line":""},{"line_number":126,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":127,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"1f621f24_96267467","line":124,"range":{"start_line":116,"start_character":0,"end_line":124,"end_character":36},"in_reply_to":"3f65232a_ceb11d92","updated":"2020-11-17 12:35:59.000000000","message":"Done","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":113,"context_line":"  💩 Almost always wrong if shared storage is in use and doesn\u0027t take"},{"line_number":114,"context_line":"  overcommit into account. Use placement."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"- ``disk_available_least``"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"  Reflects the actual disk usage on the hypervisor.  This can help identify"},{"line_number":119,"context_line":"  cases such as a host OS using more space that what is reserved for it or an"},{"line_number":120,"context_line":"  instance not consuming all of the resources allocated to it in placement."},{"line_number":121,"context_line":"  However, these cases either reflect a simple misconfiguration - a need to"},{"line_number":122,"context_line":"  increase the reserved disk space or to potentially enable disk overcommit -"},{"line_number":123,"context_line":"  or are simply useless since they have no impact on scheduling, which only"},{"line_number":124,"context_line":"  care about inventory in placement."},{"line_number":125,"context_line":""},{"line_number":126,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":127,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_ceb11d92","line":124,"range":{"start_line":116,"start_character":0,"end_line":124,"end_character":36},"in_reply_to":"3f65232a_cf095e32","updated":"2020-10-27 17:59:12.000000000","message":"actully disk_available_least is not the actual disk usage on the hypervior.\n\nit is the amount fo disk space that would be free on the hypervior if all the instance currently allocated to that hypervrios were to use all there disk  taking into account other factors like over commit and  and actual usage \n\nthis is how its calulated\n\nfree_gb \u003d total size of filesystem (minus) physical size on filesystem of all instance disks (minus) anything other than instace disks on that filesystem (e.g., /var/lib/nova/instance/_base/*)\novercommitted_total_gb \u003d virtual size of all instance disks (i.e., how big their flavor specifies) (minus) physical size on filesystem of all instance disks\ndisk_available_least \u003d free_gb - overcommitted_total_gb\n\nhttps://gist.github.com/JCallicoat/43505cab0535057ca4fb\n\nthere are cases wehre disk_available_least can be negitve which corresponde to case where you have decreased the disk allocation ration to a point were the vm currently on the host would exceed the over commit raitio if they were to use all disk. it also can be negitive if you force migitrhation too a hos bypassing the scudler(pre placment service).\n\nbut yes tl;dr its hard to use and enduser dont understand what it actully is so we really shoudl not hav ethis.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":123,"context_line":"  or are simply useless since they have no impact on scheduling, which only"},{"line_number":124,"context_line":"  care about inventory in placement."},{"line_number":125,"context_line":""},{"line_number":126,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"  Doesn\u0027t take overcommit into account. Use placement."},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_0f2b5694","line":128,"range":{"start_line":126,"start_character":0,"end_line":128,"end_character":54},"updated":"2020-10-21 16:23:13.000000000","message":"+1","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":123,"context_line":"  or are simply useless since they have no impact on scheduling, which only"},{"line_number":124,"context_line":"  care about inventory in placement."},{"line_number":125,"context_line":""},{"line_number":126,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"  Doesn\u0027t take overcommit into account. Use placement."},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_2e4ff192","line":128,"range":{"start_line":126,"start_character":0,"end_line":128,"end_character":54},"in_reply_to":"3f65232a_0f2b5694","updated":"2020-10-27 17:59:12.000000000","message":"it also does not take into account pageszie.\nit shoudl really only be repoting 4k pages but it include hugepages as free is caluated by taking the total ram and subtracting the vms current useage based on flavor. it\ndoes not correspond to the \"free\" commandline output which causes operators confusion.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"c1da65ee7f9140b77da7aa5d17dd48eec95afcc8","unresolved":false,"context_lines":[{"line_number":123,"context_line":"  or are simply useless since they have no impact on scheduling, which only"},{"line_number":124,"context_line":"  care about inventory in placement."},{"line_number":125,"context_line":""},{"line_number":126,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"  Doesn\u0027t take overcommit into account. Use placement."},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"1f621f24_d623cc54","line":128,"range":{"start_line":126,"start_character":0,"end_line":128,"end_character":54},"in_reply_to":"3f65232a_2e4ff192","updated":"2020-11-17 12:35:59.000000000","message":"Done","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"  Easily figured out by filtering running instances by host (admin-only, like"},{"line_number":133,"context_line":"  this API)."},{"line_number":134,"context_line":""},{"line_number":135,"context_line":".. [1] https://docs.openstack.org/api-ref/compute/?expanded\u003dlist-hypervisors-details-detail,show-hypervisor-details-detail"},{"line_number":136,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_aa085028","line":133,"updated":"2020-10-21 16:23:13.000000000","message":"this one is debatable. The metric is correct and we have no way to get it from Placement as we only get the resources fine-grained. Worth debating it on another round of spec, with the NUMA host topology information.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"b140fb3b680e88ffc25c46c35661b6b573aedaf7","unresolved":false,"context_lines":[{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"  Easily figured out by filtering running instances by host (admin-only, like"},{"line_number":133,"context_line":"  this API)."},{"line_number":134,"context_line":""},{"line_number":135,"context_line":".. [1] https://docs.openstack.org/api-ref/compute/?expanded\u003dlist-hypervisors-details-detail,show-hypervisor-details-detail"},{"line_number":136,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_80738580","line":133,"in_reply_to":"3f65232a_2563d1de","updated":"2020-10-26 11:09:10.000000000","message":"Actually, you\u0027re right, good point. Looks good to me to remove this field then.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"  Easily figured out by filtering running instances by host (admin-only, like"},{"line_number":133,"context_line":"  this API)."},{"line_number":134,"context_line":""},{"line_number":135,"context_line":".. [1] https://docs.openstack.org/api-ref/compute/?expanded\u003dlist-hypervisors-details-detail,show-hypervisor-details-detail"},{"line_number":136,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_ee54d923","line":133,"in_reply_to":"3f65232a_80738580","updated":"2020-10-27 17:59:12.000000000","message":"yep is basiclaly that to wc\n\nsoemthing like \"openstack server list --host $HOST -f value -c UUID | wc -l\"","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"8ec36c4ec8a11fd1d6c02f308d50cb06deb945ec","unresolved":false,"context_lines":[{"line_number":130,"context_line":"- ``running_vms``"},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"  Easily figured out by filtering running instances by host (admin-only, like"},{"line_number":133,"context_line":"  this API)."},{"line_number":134,"context_line":""},{"line_number":135,"context_line":".. [1] https://docs.openstack.org/api-ref/compute/?expanded\u003dlist-hypervisors-details-detail,show-hypervisor-details-detail"},{"line_number":136,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_2563d1de","line":133,"in_reply_to":"3f65232a_aa085028","updated":"2020-10-21 17:37:30.000000000","message":"Isn\u0027t\n\n  openstack server list --host $HOST\n\nGood enough? It\u0027s not the exact same as \u0027running_vms\u0027 will show VMs not under nova\u0027s control, but that\u0027s considered a misconfiguration afaik?","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":137,"context_line":"Use Cases"},{"line_number":138,"context_line":"---------"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"As a user, I don\u0027t want to see misleading information reported from my API."},{"line_number":141,"context_line":""},{"line_number":142,"context_line":"Proposed change"},{"line_number":143,"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":"3f65232a_2a1c4069","line":140,"updated":"2020-10-21 16:23:13.000000000","message":"there, I agree. For things we know that suck, kill them. But you\u0027re also packing things that are not replaced, and for that, I disagree.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":137,"context_line":"Use Cases"},{"line_number":138,"context_line":"---------"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"As a user, I don\u0027t want to see misleading information reported from my API."},{"line_number":141,"context_line":""},{"line_number":142,"context_line":"Proposed change"},{"line_number":143,"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":"3f65232a_8e3ba5e9","line":140,"in_reply_to":"3f65232a_2a1c4069","updated":"2020-10-27 17:59:12.000000000","message":"well we are proposing deprecating all of os-hypervirors and removing it entirly after walaby so i see this as a fallback propal for the minium set of things we shoudl remove if we dont deprecate/remove the entire api.","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"c9054137550c6bea49f90811edb45fc4b846830f","unresolved":false,"context_lines":[{"line_number":166,"context_line":"``free_ram_mb``, ``memory_mb``, ``memory_mb_used`` and ``running_vms``."},{"line_number":167,"context_line":""},{"line_number":168,"context_line":"Starting from the new API microversion, the ``/os-hypervisors/statistics`` API"},{"line_number":169,"context_line":"will be removed entirely and will return a HTTP 410 (Gone)."},{"line_number":170,"context_line":""},{"line_number":171,"context_line":"Security impact"},{"line_number":172,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_aa2170a2","line":169,"updated":"2020-10-21 16:23:13.000000000","message":"+1000","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"e4ae9628256945767040301bd4f8e973b5fab27c","unresolved":false,"context_lines":[{"line_number":166,"context_line":"``free_ram_mb``, ``memory_mb``, ``memory_mb_used`` and ``running_vms``."},{"line_number":167,"context_line":""},{"line_number":168,"context_line":"Starting from the new API microversion, the ``/os-hypervisors/statistics`` API"},{"line_number":169,"context_line":"will be removed entirely and will return a HTTP 410 (Gone)."},{"line_number":170,"context_line":""},{"line_number":171,"context_line":"Security impact"},{"line_number":172,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f65232a_4e1a2d7d","line":169,"in_reply_to":"3f65232a_aa2170a2","updated":"2020-10-27 17:59:12.000000000","message":"+1","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"00976a2ae30b20db329318bde27fc9fef36e003f","unresolved":false,"context_lines":[{"line_number":207,"context_line":"  ``cpu_info.topology`` values are not relevant for scheduling and therefore"},{"line_number":208,"context_line":"  have no direct replacement in placement or another API. They can, however, be"},{"line_number":209,"context_line":"  identified through inspection of the host."},{"line_number":210,"context_line":""},{"line_number":211,"context_line":"Performance Impact"},{"line_number":212,"context_line":"------------------"},{"line_number":213,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"1f621f24_b293a9a3","line":210,"updated":"2020-10-28 16:42:40.000000000","message":"Horizon will also need to be updated","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"c1da65ee7f9140b77da7aa5d17dd48eec95afcc8","unresolved":false,"context_lines":[{"line_number":207,"context_line":"  ``cpu_info.topology`` values are not relevant for scheduling and therefore"},{"line_number":208,"context_line":"  have no direct replacement in placement or another API. They can, however, be"},{"line_number":209,"context_line":"  identified through inspection of the host."},{"line_number":210,"context_line":""},{"line_number":211,"context_line":"Performance Impact"},{"line_number":212,"context_line":"------------------"},{"line_number":213,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"1f621f24_36d84051","line":210,"in_reply_to":"1f621f24_b293a9a3","updated":"2020-11-17 12:35:59.000000000","message":"Done","commit_id":"92bbfec70ea08c1c1aaf4bfcc696bf975a2082c9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"7373c4ea5a45aff265885dde5463a8307de20396","unresolved":false,"context_lines":[{"line_number":101,"context_line":"  \"will be equal or greater than the number of active VMs on the system (it can"},{"line_number":102,"context_line":"  be greater when VMs are being deleted and the hypervisor is still cleaning"},{"line_number":103,"context_line":"  up)\" [2]_. This information is easily calculated by listing active and"},{"line_number":104,"context_line":"  deleted instances."},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"- ``cpu_info``"},{"line_number":107,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_805aa97f","line":104,"updated":"2020-11-17 14:59:41.000000000","message":"++, we agreed on removing it at the PTG.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"8e8c620e3f791e1397bf2e27323601468b6a35f1","unresolved":false,"context_lines":[{"line_number":108,"context_line":"  Useful at face value but the only thing relevant for scheduling purposes are"},{"line_number":109,"context_line":"  the CPU architecture and CPU features, all of which are already handled by"},{"line_number":110,"context_line":"  placement trait requests. The topology field is an oddity that should likely"},{"line_number":111,"context_line":"  never have been added. It\u0027s not usable in scheduling and is possibly wrong,"},{"line_number":112,"context_line":"  given it doesn\u0027t reflect offline CPUs or those not available to nova due to"},{"line_number":113,"context_line":"  configuration, and it doesn\u0027t handle non-uniform CPU topologies where there"},{"line_number":114,"context_line":"  are e.g. more cores on one socket than another. If the operator wants this"}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_2a68d4f2","line":111,"range":{"start_line":111,"start_character":59,"end_line":111,"end_character":77},"updated":"2020-11-17 16:22:01.000000000","message":"it is used to decide if a target host is acceptable for live migration. So I hope this data is not wrong :) But overall I\u0027m OK not to expose this as the format is not define in the o.vo level at all.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"7373c4ea5a45aff265885dde5463a8307de20396","unresolved":false,"context_lines":[{"line_number":113,"context_line":"  configuration, and it doesn\u0027t handle non-uniform CPU topologies where there"},{"line_number":114,"context_line":"  are e.g. more cores on one socket than another. If the operator wants this"},{"line_number":115,"context_line":"  information, they can simply inspect the host like they would have to do to"},{"line_number":116,"context_line":"  identify e.g. the specifics of PCI devices or storage devices."},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"- ``free_disk_gb``, ``local_gb``, ``local_gb_used``"},{"line_number":119,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_605dcd77","line":116,"updated":"2020-11-17 14:59:41.000000000","message":"we agreed on removing it too at the PTG since this blob doesn\u0027t really help.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"7373c4ea5a45aff265885dde5463a8307de20396","unresolved":false,"context_lines":[{"line_number":126,"context_line":"  instances on the host were to use all their allocated disk. This can go"},{"line_number":127,"context_line":"  negative if disk overcommit is enabled or if an instance is force migrated to"},{"line_number":128,"context_line":"  a host, bypassing the scheduler. This value is hard to use and frequently"},{"line_number":129,"context_line":"  misunderstood by end-users. Use placement."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_004e99bf","line":129,"updated":"2020-11-17 14:59:41.000000000","message":"++, thanks for modifying it.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"8e8c620e3f791e1397bf2e27323601468b6a35f1","unresolved":false,"context_lines":[{"line_number":126,"context_line":"  instances on the host were to use all their allocated disk. This can go"},{"line_number":127,"context_line":"  negative if disk overcommit is enabled or if an instance is force migrated to"},{"line_number":128,"context_line":"  a host, bypassing the scheduler. This value is hard to use and frequently"},{"line_number":129,"context_line":"  misunderstood by end-users. Use placement."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"- ``free_ram_mb``, ``memory_mb``, ``memory_mb_used``"},{"line_number":132,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_9e7a0621","line":129,"range":{"start_line":129,"start_character":29,"end_line":129,"end_character":44},"updated":"2020-11-17 16:22:01.000000000","message":"Drop this. Placement is not a replacement for this as it does not have this information. Here the admin needs to go and inspect the hypervisor directly to get the information.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"b1f0c8a6f7120820a876773b79279dcdeb6f8cda","unresolved":false,"context_lines":[{"line_number":151,"context_line":"Remove the resource-related fields from the output of the"},{"line_number":152,"context_line":"``/os-hypervisors/detail`` API and remove the ``/os-hypervisors/statistics``"},{"line_number":153,"context_line":"API in its entirety."},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Alternatives"},{"line_number":156,"context_line":"------------"},{"line_number":157,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_23a63fc2","line":154,"updated":"2020-11-17 15:40:08.000000000","message":"so i would like to add something there too that we discussed breifly on irc.\n\nnamely the idea that we shoudl allow users with the PROECT_ADMIN role to list all hyperviors there project is allowed to acess via the summary endpoint\n\n\nbasically we would use the aggreate metaddata used for the tenat affinity (filter or with placment) to retrun what hosts they can see the summary of.\n\nif there is not aggreate medata for the tenatn then they can see all host on the system if there is then they see the filtered view.\nthis will mean we dont have to give them system_reader for any of the exiting users like boot to a specific host.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"78588bf50c06fac115611e5864e93d05e54aeb18","unresolved":false,"context_lines":[{"line_number":151,"context_line":"Remove the resource-related fields from the output of the"},{"line_number":152,"context_line":"``/os-hypervisors/detail`` API and remove the ``/os-hypervisors/statistics``"},{"line_number":153,"context_line":"API in its entirety."},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Alternatives"},{"line_number":156,"context_line":"------------"},{"line_number":157,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_be40a2a6","line":154,"in_reply_to":"1f621f24_23a63fc2","updated":"2020-11-17 16:34:39.000000000","message":"but hypervisor is not project level info right? I mean if project can see all the host then we goes back to legacy rbac. Main idea of new policies are to strict default for sec purpose.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"9fc7bed4836882401082ddb901ce9c04a508bbdc","unresolved":false,"context_lines":[{"line_number":151,"context_line":"Remove the resource-related fields from the output of the"},{"line_number":152,"context_line":"``/os-hypervisors/detail`` API and remove the ``/os-hypervisors/statistics``"},{"line_number":153,"context_line":"API in its entirety."},{"line_number":154,"context_line":""},{"line_number":155,"context_line":"Alternatives"},{"line_number":156,"context_line":"------------"},{"line_number":157,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_be570258","line":154,"in_reply_to":"1f621f24_be40a2a6","updated":"2020-11-17 16:38:43.000000000","message":"Let\u0027s debate this separately. I\u0027ve proposed an amendment [1] to do just that\n\n[1] https://review.opendev.org/763043","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"7373c4ea5a45aff265885dde5463a8307de20396","unresolved":false,"context_lines":[{"line_number":271,"context_line":"      | HW_CPU_X86_SVM                        |"},{"line_number":272,"context_line":"      | HW_CPU_HYPERTHREADING                 |"},{"line_number":273,"context_line":"      | HW_CPU_X86_MMX                        |"},{"line_number":274,"context_line":""},{"line_number":275,"context_line":"- The ``disk_available_least``, ``cpu_info.model``, ``cpu_info.vendor`` and"},{"line_number":276,"context_line":"  ``cpu_info.topology`` values are not relevant for scheduling and therefore"},{"line_number":277,"context_line":"  have no direct replacement in placement or another API. They can, however, be"}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_a0d7e5be","line":274,"updated":"2020-11-17 14:59:41.000000000","message":"thanks for the alternatives, good for documenting them.","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"7373c4ea5a45aff265885dde5463a8307de20396","unresolved":false,"context_lines":[{"line_number":278,"context_line":"  identified through inspection of the host."},{"line_number":279,"context_line":""},{"line_number":280,"context_line":"Horizon will need to be updated to talk to placement or use this API with an"},{"line_number":281,"context_line":"older microversion."},{"line_number":282,"context_line":""},{"line_number":283,"context_line":"Performance Impact"},{"line_number":284,"context_line":"------------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"1f621f24_60d2edd2","line":281,"updated":"2020-11-17 14:59:41.000000000","message":"good luck with communicating on this, I guess you won\u0027t do it, right? ;)","commit_id":"e7876cd57c9134fc0b45986bcc6be53ccdd41a81"}]}