)]}'
{"placement-api-ref/source/aggregates.inc":[{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"652e25c04f3693e2416e6609925345b9ed7462cd","unresolved":false,"context_lines":[{"line_number":8,"context_line":"with one resource provider."},{"line_number":9,"context_line":""},{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_143608b6","line":11,"updated":"2018-02-15 16:30:53.000000000","message":"Maybe to this note we need to add:\n\n\"However, Placement aggregates could be considered a superset of Nova host aggregates in order to mirror a grouping of compute providers in Placement. That does not preclude their usage for other services to group resource providers as they see, like IP allocation pools with the networking service.\"","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":1063,"name":"Ed Leafe","email":"ed@leafe.com","username":"ed-leafe"},"change_message_id":"e097edef092199a6f08a8e8a847daa836af6370d","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_4b5df1e7","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"updated":"2018-02-14 17:34:56.000000000","message":"This a too constraining. There may be other uses for placement aggregates in the future, such as affinity, that don\u0027t involve sharing resources. Perhaps just saying \"are currently used\" to avoid precluding other future uses would be better.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":1063,"name":"Ed Leafe","email":"ed@leafe.com","username":"ed-leafe"},"change_message_id":"721d8dde01de96d8512f604e2f97e09b5fe63794","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_2f88916f","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_142da845","updated":"2018-02-15 17:01:08.000000000","message":"The highlighted sentence is still incorrect. The compute nodes in a mirrored nova host aggregate do not share resources. They simply have *some* sort of relationship.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"0af46fe3e15814ad1619ff5611ce6e1483cbc1bb","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_ef44e91d","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_2f88916f","updated":"2018-02-15 17:03:43.000000000","message":"Yep, don\u0027t disagree with that, just the \"identical\" assertion.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":1063,"name":"Ed Leafe","email":"ed@leafe.com","username":"ed-leafe"},"change_message_id":"369b6607397d2070773adaaa93e98544b3c69004","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_d9179445","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_4b5df1e7","updated":"2018-02-14 20:09:47.000000000","message":"OK, now with Dan\u0027s comments [0], it seems that there is intent to have all Nova aggregates mirrored into placement, so I think that this paragraph and even the note above it are not accurate. I realize that this is not yet implemented, but it would be misleading to make a statement like this when we are actively working on making it false. It should be more like: \"Placement aggregates are identical to Nova aggregates, with the exception that there may be some Placement-specific aggregates defined for shared resources. An example...\"\n\n[0] https://review.openstack.org/#/c/544585/2/specs/rocky/approved/placement-req-filter.rst@89","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":14070,"name":"Eric Fried","email":"openstack@fried.cc","username":"efried"},"change_message_id":"cb3ce4b502d24cb1ffbb077be8793bcbf0518c95","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_7c2080f8","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_7bf40d5c","updated":"2018-02-16 14:58:42.000000000","message":"Followup here: https://review.openstack.org/545356","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"e8bfb4360ea69d72528f7605119e5066223650dd","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_7bf40d5c","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_8f20c54b","updated":"2018-02-16 09:44:07.000000000","message":"We can discuss a lot about that, but I\u0027m fine with merging that doc now, and have a follow-up in case people still argue.\n\nIt\u0027s doc, not code.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2322312f4d15449fb1711b6f4b2d611bc62cee6f","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_142da845","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_d9179445","updated":"2018-02-15 16:32:47.000000000","message":"Just because nova auto-creates aggregates in placement doesn\u0027t make them \"identical\". Cinder could/would do the same, which makes placement a superset of the grouping constructs of its users. We could also only choose to mirror aggregates up to placement that have certain metadata items indicating that they\u0027re relevant to placement.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"ea3ad2279877212fc9fdc3bcd79b41e8e9e875be","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_14d12858","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_d9179445","updated":"2018-02-15 16:28:37.000000000","message":"Placement aggregates aren\u0027t identical to nova host aggregates though, for one thing they don\u0027t have nor will have metadata.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"},{"author":{"_account_id":1063,"name":"Ed Leafe","email":"ed@leafe.com","username":"ed-leafe"},"change_message_id":"796a4c37145d6fbd13c63dea829826c51118c66c","unresolved":false,"context_lines":[{"line_number":10,"context_line":".. note:: Placement aggregates are *not* the same as Nova host"},{"line_number":11,"context_line":"   aggregates and should not be considered equivalent."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Provider aggregates are used for modeling relationships where a provider"},{"line_number":14,"context_line":"shares resources with one or more other providers. An example is a shared"},{"line_number":15,"context_line":"storage pool providing DISK_GB resources to other compute node providers that"},{"line_number":16,"context_line":"provide VCPU and MEMORY_MB resources."},{"line_number":17,"context_line":""}],"source_content_type":"text/x-c++src","patch_set":1,"id":"3fa0c359_8f20c54b","line":14,"range":{"start_line":13,"start_character":0,"end_line":14,"end_character":50},"in_reply_to":"3fa0c359_ef44e91d","updated":"2018-02-15 17:05:45.000000000","message":"Of course; \"superset\" would be fine, too. Just some wording to acknowledge the sync between them.","commit_id":"03be4beb4c122cee99e91cc78435e7ba38556fa1"}]}