)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"656f0f8153cad6ff5a4fbeca71befceb77c7a952","unresolved":false,"context_lines":[{"line_number":7,"context_line":"Update host_state\u0027s limits when specified AZ"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"- Create vms where specifies force_hosts, if request resource"},{"line_number":10,"context_line":"  exceeds the total resource, the vms are spawned succeefully"},{"line_number":11,"context_line":"  without error occurred, that was abnormal because resource"},{"line_number":12,"context_line":"  filters(such as ram filter, disk filter) were specified."},{"line_number":13,"context_line":"- In this patch, When _schedule, update host\u0027s limits if filters"},{"line_number":14,"context_line":"  specified. Spawning vms would be failed if requested resources"},{"line_number":15,"context_line":"  were bigger than the usable resource."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":3,"id":"ffb9cba7_71d755ef","line":12,"range":{"start_line":10,"start_character":30,"end_line":12,"end_character":58},"updated":"2019-04-23 21:57:36.000000000","message":"This doesn\u0027t make sense with the first part of this sentence which says \"Create vms where specifies force_hosts\" - if you\u0027re forcing the host, you\u0027re bypassing the filters so there is no filtering (or ResourceTracker claim) on those resources (cores/ram/disk) outside of the placement service (since pike anyway).","commit_id":"ed02d5db1bf94dee2e5d78b89abed85500656347"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"656f0f8153cad6ff5a4fbeca71befceb77c7a952","unresolved":false,"context_lines":[{"line_number":10,"context_line":"  exceeds the total resource, the vms are spawned succeefully"},{"line_number":11,"context_line":"  without error occurred, that was abnormal because resource"},{"line_number":12,"context_line":"  filters(such as ram filter, disk filter) were specified."},{"line_number":13,"context_line":"- In this patch, When _schedule, update host\u0027s limits if filters"},{"line_number":14,"context_line":"  specified. Spawning vms would be failed if requested resources"},{"line_number":15,"context_line":"  were bigger than the usable resource."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Change-Id: Ib74599d123ceb1acffa68a6192e6635c878cfa0e"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":3,"id":"ffb9cba7_b1f62d8f","line":14,"range":{"start_line":13,"start_character":54,"end_line":14,"end_character":11},"updated":"2019-04-23 21:57:36.000000000","message":"What does this mean? If filters are enabled which populate the limits dict which goes to the ResourceTracker.instance_claim? As noted, if you\u0027re forcing the host you\u0027re bypassing the filters and risking over-subscription of resources on the forced host. That\u0027s why we have this spec in Train:\n\nhttps://specs.openstack.org/openstack/nova-specs/specs/train/approved/add-host-and-hypervisor-hostname-flag-to-create-server.html\n\nSo with that ^ you can create a server on a requested host but it\u0027s still filtered in the scheduler (which would populate limits as well).","commit_id":"ed02d5db1bf94dee2e5d78b89abed85500656347"}],"nova/scheduler/filters/__init__.py":[{"author":{"_account_id":7,"name":"Jay Pipes","email":"jaypipes@gmail.com","username":"jaypipes"},"change_message_id":"5b383a712e8c466ca252aba7e7c3f72efadba988","unresolved":false,"context_lines":[{"line_number":51,"context_line":""},{"line_number":52,"context_line":"    def host_limits_update(self, host_state, filter_properties):"},{"line_number":53,"context_line":"        \"\"\"Update the host_state.limits parameter. Override this in a"},{"line_number":54,"context_line":"        subclass, if the subclass does need to update, just pass."},{"line_number":55,"context_line":"        \"\"\""},{"line_number":56,"context_line":"        pass"},{"line_number":57,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"dfd5e7cf_9c120409","line":54,"range":{"start_line":54,"start_character":33,"end_line":54,"end_character":43},"updated":"2019-01-09 19:05:55.000000000","message":"does not need, you mean?","commit_id":"ed02d5db1bf94dee2e5d78b89abed85500656347"}],"nova/scheduler/filters/core_filter.py":[{"author":{"_account_id":7,"name":"Jay Pipes","email":"jaypipes@gmail.com","username":"jaypipes"},"change_message_id":"5b383a712e8c466ca252aba7e7c3f72efadba988","unresolved":false,"context_lines":[{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"class CoreFilter(BaseCoreFilter):"},{"line_number":87,"context_line":"    \"\"\"DEPRECATED: CoreFilter filters based on CPU core utilization.\"\"\""},{"line_number":88,"context_line":""},{"line_number":89,"context_line":"    def __init__(self):"},{"line_number":90,"context_line":"        super(CoreFilter, self).__init__()"}],"source_content_type":"text/x-python","patch_set":3,"id":"dfd5e7cf_9ce0c4fc","line":87,"range":{"start_line":87,"start_character":4,"end_line":87,"end_character":71},"updated":"2019-01-09 19:05:55.000000000","message":"did you notice this? :)\n\nThe filters for RAM, vCPU and disk have been deprecated because we use the placement service now for handling resource requests for these basic resource classes.","commit_id":"ed02d5db1bf94dee2e5d78b89abed85500656347"}]}