)]}'
{"rally-jobs/neutron-neutron.yaml":[{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"00a02303809520f64ac75d6fd581d900e772b0f4","unresolved":false,"context_lines":[{"line_number":66,"context_line":"      args:"},{"line_number":67,"context_line":"        network_create_args:"},{"line_number":68,"context_line":"        port_create_args:"},{"line_number":69,"context_line":"        ports_per_network: 200"},{"line_number":70,"context_line":"      runner:"},{"line_number":71,"context_line":"        type: \"constant\""},{"line_number":72,"context_line":"        times: 40"}],"source_content_type":"text/x-yaml","patch_set":2,"id":"3ac371cc_5e8d957a","line":69,"updated":"2016-08-14 07:36:06.000000000","message":"can you please clarify why 200 ports per network is a real simulation of a system.\nI have seen the following use cases:\n1. a handful of ports per network\n2. between 40 - 50 ports per network - this is with people using pretty elaborate heat stack deploying topologies\n3. extreme cases when a user has a few flat networks and they have thousands of ports on the same network","commit_id":"a16f8ce18ebc1cb10fba570e5786b8190982c787"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"e7b3797b656c41a11d7189c6be1565f478e065c2","unresolved":false,"context_lines":[{"line_number":66,"context_line":"      args:"},{"line_number":67,"context_line":"        network_create_args:"},{"line_number":68,"context_line":"        port_create_args:"},{"line_number":69,"context_line":"        ports_per_network: 200"},{"line_number":70,"context_line":"      runner:"},{"line_number":71,"context_line":"        type: \"constant\""},{"line_number":72,"context_line":"        times: 40"}],"source_content_type":"text/x-yaml","patch_set":2,"id":"3ac371cc_d9b87fe8","line":69,"in_reply_to":"3ac371cc_5e8d957a","updated":"2016-08-14 09:58:34.000000000","message":"Well this many ports per network would automatically cover the first two cases. If the system can handle more ports, it can obviously handle fewer.\n\nI haven\u0027t seen the flat networks case exceeding a thousand. Maybe we can add another test case for that. I had just picked 200 because that\u0027s a rough approximation of vlan-based ones I have seen.","commit_id":"a16f8ce18ebc1cb10fba570e5786b8190982c787"},{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"00a02303809520f64ac75d6fd581d900e772b0f4","unresolved":false,"context_lines":[{"line_number":80,"context_line":"             network: -1"},{"line_number":81,"context_line":"             subnet: -1"},{"line_number":82,"context_line":"             router: -1"},{"line_number":83,"context_line":"             port: 4000  # ports per network * conurrency"},{"line_number":84,"context_line":"      sla:"},{"line_number":85,"context_line":"        max_avg_duration_per_atomic:"},{"line_number":86,"context_line":"          neutron.list_ports: 5"}],"source_content_type":"text/x-yaml","patch_set":2,"id":"3ac371cc_9eed4d28","line":83,"updated":"2016-08-14 07:36:06.000000000","message":"why do we need to limit the quota here?","commit_id":"a16f8ce18ebc1cb10fba570e5786b8190982c787"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"e7b3797b656c41a11d7189c6be1565f478e065c2","unresolved":false,"context_lines":[{"line_number":80,"context_line":"             network: -1"},{"line_number":81,"context_line":"             subnet: -1"},{"line_number":82,"context_line":"             router: -1"},{"line_number":83,"context_line":"             port: 4000  # ports per network * conurrency"},{"line_number":84,"context_line":"      sla:"},{"line_number":85,"context_line":"        max_avg_duration_per_atomic:"},{"line_number":86,"context_line":"          neutron.list_ports: 5"}],"source_content_type":"text/x-yaml","patch_set":2,"id":"3ac371cc_19cb3774","line":83,"in_reply_to":"3ac371cc_9eed4d28","updated":"2016-08-14 09:58:34.000000000","message":"By having a -1 quota, we effectively disable the quota engine so it doesn\u0027t actually exercise the performance of the quota engine. By enabling it we ensure that we don\u0027t get failures when the quota engine is doing its job.","commit_id":"a16f8ce18ebc1cb10fba570e5786b8190982c787"}]}