)]}'
{"neutron/agent/metadata/agent.py":[{"author":{"_account_id":11975,"name":"Slawek Kaplonski","email":"skaplons@redhat.com","username":"slaweq"},"change_message_id":"1c2a610772b70dd00db31c2192c6ae068ee7f431","unresolved":false,"context_lines":[{"line_number":365,"context_line":"    def run(self):"},{"line_number":366,"context_line":"        server \u003d agent_utils.UnixDomainWSGIServer("},{"line_number":367,"context_line":"                     constants.AGENT_PROCESS_METADATA)"},{"line_number":368,"context_line":"        # Set the default metadata_workers for ML2/OVS, if needed"},{"line_number":369,"context_line":"        md_workers \u003d self.conf.metadata_workers"},{"line_number":370,"context_line":"        if md_workers is None:"},{"line_number":371,"context_line":"            md_workers \u003d host.cpu_count() // 2"}],"source_content_type":"text/x-python","patch_set":3,"id":"9f560f44_ccf54a53","line":368,"range":{"start_line":368,"start_character":51,"end_line":368,"end_character":54},"updated":"2020-09-08 20:37:21.000000000","message":"it\u0027s not only OVS :)","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"34edbd45b36351629d97eb9e159dac118c734993","unresolved":false,"context_lines":[{"line_number":365,"context_line":"    def run(self):"},{"line_number":366,"context_line":"        server \u003d agent_utils.UnixDomainWSGIServer("},{"line_number":367,"context_line":"                     constants.AGENT_PROCESS_METADATA)"},{"line_number":368,"context_line":"        # Set the default metadata_workers for ML2/OVS, if needed"},{"line_number":369,"context_line":"        md_workers \u003d self.conf.metadata_workers"},{"line_number":370,"context_line":"        if md_workers is None:"},{"line_number":371,"context_line":"            md_workers \u003d host.cpu_count() // 2"}],"source_content_type":"text/x-python","patch_set":3,"id":"9f560f44_55d6d6f3","line":368,"range":{"start_line":368,"start_character":51,"end_line":368,"end_character":54},"in_reply_to":"9f560f44_ccf54a53","updated":"2020-09-09 08:14:55.000000000","message":"++","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"}],"neutron/agent/ovn/metadata/server.py":[{"author":{"_account_id":23804,"name":"Daniel Alvarez","email":"dalvarez@redhat.com","username":"dalvarez"},"change_message_id":"c8e45134f6314959ac5cdaafb106d0cd12980b76","unresolved":false,"context_lines":[{"line_number":202,"context_line":"        # Set the default metadata_workers for ML2/OVN, if needed"},{"line_number":203,"context_line":"        md_workers \u003d self.conf.metadata_workers"},{"line_number":204,"context_line":"        if md_workers is None:"},{"line_number":205,"context_line":"            md_workers \u003d 2"},{"line_number":206,"context_line":"        self.server.start(MetadataProxyHandler(self.conf),"},{"line_number":207,"context_line":"                          self.conf.metadata_proxy_socket,"},{"line_number":208,"context_line":"                          workers\u003dmd_workers,"}],"source_content_type":"text/x-python","patch_set":2,"id":"9f560f44_0567274f","line":205,"range":{"start_line":205,"start_character":0,"end_line":205,"end_character":26},"updated":"2020-09-08 08:30:32.000000000","message":"What about having two different options? I understand this and the rationale but if a deployer sets the default somehow based on the conf, this is not that obvious. Also it\u0027s kind of breaking the model for the different config options and their defaults.","commit_id":"1c9cc84fead5be3dedc83d9394b0d8c236d96ab3"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"48fce73fe88b66df64a195beb28f5b31f041b271","unresolved":false,"context_lines":[{"line_number":202,"context_line":"        # Set the default metadata_workers for ML2/OVN, if needed"},{"line_number":203,"context_line":"        md_workers \u003d self.conf.metadata_workers"},{"line_number":204,"context_line":"        if md_workers is None:"},{"line_number":205,"context_line":"            md_workers \u003d 2"},{"line_number":206,"context_line":"        self.server.start(MetadataProxyHandler(self.conf),"},{"line_number":207,"context_line":"                          self.conf.metadata_proxy_socket,"},{"line_number":208,"context_line":"                          workers\u003dmd_workers,"}],"source_content_type":"text/x-python","patch_set":2,"id":"9f560f44_581e225f","line":205,"range":{"start_line":205,"start_character":0,"end_line":205,"end_character":26},"in_reply_to":"9f560f44_0567274f","updated":"2020-09-08 08:48:53.000000000","message":"A default somehow based on the conf ? If the deployer sets any value to this conf it will be honored, cause this is only set if the metadata_workers config is unset.\n\nYeah the multiple defaults approach for this is not conventional, tho, the default for this option wasn\u0027t static to begin with, it was always calculated on the fly (based on # of CPUs). The difference is that now it also takes in consideration the backend driver being used, it\u0027s not that off/different I would argue.\n\nIf we go with the two different options approach we will end up breaking some deployment tools ways of setting this option for ML2/OVN as they will need to add code to make use of the new option.\n\nWhat you think ?","commit_id":"1c9cc84fead5be3dedc83d9394b0d8c236d96ab3"},{"author":{"_account_id":23804,"name":"Daniel Alvarez","email":"dalvarez@redhat.com","username":"dalvarez"},"change_message_id":"fda1049eef51f0a535f21f78beab65e95a27aa9b","unresolved":false,"context_lines":[{"line_number":202,"context_line":"        # Set the default metadata_workers for ML2/OVN, if needed"},{"line_number":203,"context_line":"        md_workers \u003d self.conf.metadata_workers"},{"line_number":204,"context_line":"        if md_workers is None:"},{"line_number":205,"context_line":"            md_workers \u003d 2"},{"line_number":206,"context_line":"        self.server.start(MetadataProxyHandler(self.conf),"},{"line_number":207,"context_line":"                          self.conf.metadata_proxy_socket,"},{"line_number":208,"context_line":"                          workers\u003dmd_workers,"}],"source_content_type":"text/x-python","patch_set":2,"id":"9f560f44_f8d73622","line":205,"range":{"start_line":205,"start_character":0,"end_line":205,"end_character":26},"in_reply_to":"9f560f44_581e225f","updated":"2020-09-08 09:05:14.000000000","message":"Fair enough :)","commit_id":"1c9cc84fead5be3dedc83d9394b0d8c236d96ab3"}],"neutron/conf/agent/metadata/config.py":[{"author":{"_account_id":11975,"name":"Slawek Kaplonski","email":"skaplons@redhat.com","username":"slaweq"},"change_message_id":"1c2a610772b70dd00db31c2192c6ae068ee7f431","unresolved":false,"context_lines":[{"line_number":95,"context_line":"               sample_default\u003d\u0027\u003cnum_of_cpus\u003e / 2\u0027,"},{"line_number":96,"context_line":"               help\u003d_(\u0027Number of separate worker processes for metadata \u0027"},{"line_number":97,"context_line":"                      \u0027server (defaults to half of the number of CPUs when \u0027"},{"line_number":98,"context_line":"                      \u0027used with ML2/OVS or 2 when used with ML2/OVN)\u0027)),"},{"line_number":99,"context_line":"    cfg.IntOpt(\u0027metadata_backlog\u0027,"},{"line_number":100,"context_line":"               default\u003d4096,"},{"line_number":101,"context_line":"               help\u003d_(\u0027Number of backlog requests to configure the \u0027"}],"source_content_type":"text/x-python","patch_set":3,"id":"9f560f44_8cffd232","line":98,"range":{"start_line":98,"start_character":23,"end_line":98,"end_character":68},"updated":"2020-09-08 20:37:21.000000000","message":"I would say \"2 when used with ML2/OVN and half of the number of CPUs with other backend drivers\" or something like that becuase it\u0027s not only ML2/OVS but can be also e.g. Linuxbridge.","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"34edbd45b36351629d97eb9e159dac118c734993","unresolved":false,"context_lines":[{"line_number":95,"context_line":"               sample_default\u003d\u0027\u003cnum_of_cpus\u003e / 2\u0027,"},{"line_number":96,"context_line":"               help\u003d_(\u0027Number of separate worker processes for metadata \u0027"},{"line_number":97,"context_line":"                      \u0027server (defaults to half of the number of CPUs when \u0027"},{"line_number":98,"context_line":"                      \u0027used with ML2/OVS or 2 when used with ML2/OVN)\u0027)),"},{"line_number":99,"context_line":"    cfg.IntOpt(\u0027metadata_backlog\u0027,"},{"line_number":100,"context_line":"               default\u003d4096,"},{"line_number":101,"context_line":"               help\u003d_(\u0027Number of backlog requests to configure the \u0027"}],"source_content_type":"text/x-python","patch_set":3,"id":"9f560f44_15d0def6","line":98,"range":{"start_line":98,"start_character":23,"end_line":98,"end_character":68},"in_reply_to":"9f560f44_8cffd232","updated":"2020-09-09 08:14:55.000000000","message":"Ah great point! ++","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"}],"releasenotes/notes/ovn-metadata_workers-1121334593ce9829.yaml":[{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"3777d081079bb02a0f6e673852b795eec7ef6434","unresolved":false,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"upgrade:"},{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    The default value for the \"metadata_workers\" configuration option"},{"line_number":5,"context_line":"    has changed to 2 for the ML2/OVN driver. For ML2/OVS the default"},{"line_number":6,"context_line":"    value remains the same. Each driver has different approaches when"},{"line_number":7,"context_line":"    serving metadata to the instances and the previous default value of"}],"source_content_type":"text/x-yaml","patch_set":3,"id":"9f560f44_1d312fee","line":4,"range":{"start_line":4,"start_character":30,"end_line":4,"end_character":48},"updated":"2020-09-08 18:17:44.000000000","message":"s/``metadata_workers``\n\nwhich will make it bold in the relnote","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"34edbd45b36351629d97eb9e159dac118c734993","unresolved":false,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"upgrade:"},{"line_number":3,"context_line":"  - |"},{"line_number":4,"context_line":"    The default value for the \"metadata_workers\" configuration option"},{"line_number":5,"context_line":"    has changed to 2 for the ML2/OVN driver. For ML2/OVS the default"},{"line_number":6,"context_line":"    value remains the same. Each driver has different approaches when"},{"line_number":7,"context_line":"    serving metadata to the instances and the previous default value of"}],"source_content_type":"text/x-yaml","patch_set":3,"id":"9f560f44_b5e772e3","line":4,"range":{"start_line":4,"start_character":30,"end_line":4,"end_character":48},"in_reply_to":"9f560f44_1d312fee","updated":"2020-09-09 08:14:55.000000000","message":"Done","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"3777d081079bb02a0f6e673852b795eec7ef6434","unresolved":false,"context_lines":[{"line_number":5,"context_line":"    has changed to 2 for the ML2/OVN driver. For ML2/OVS the default"},{"line_number":6,"context_line":"    value remains the same. Each driver has different approaches when"},{"line_number":7,"context_line":"    serving metadata to the instances and the previous default value of"},{"line_number":8,"context_line":"    \"\u003cnumber of CPUs\u003e / 2\" did not make sense for ML2/OVN as the ovn"},{"line_number":9,"context_line":"    metadata agents are distributed running on Compute nodes instead of"},{"line_number":10,"context_line":"    Controller nodes. In fact, the previous default value could cause"},{"line_number":11,"context_line":"    scalability issues with ML2/OVN and was overwritten by the deployment"}],"source_content_type":"text/x-yaml","patch_set":3,"id":"9f560f44_7d22eba4","line":8,"range":{"start_line":8,"start_character":65,"end_line":8,"end_character":68},"updated":"2020-09-08 18:17:44.000000000","message":"nit: OVN","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"34edbd45b36351629d97eb9e159dac118c734993","unresolved":false,"context_lines":[{"line_number":5,"context_line":"    has changed to 2 for the ML2/OVN driver. For ML2/OVS the default"},{"line_number":6,"context_line":"    value remains the same. Each driver has different approaches when"},{"line_number":7,"context_line":"    serving metadata to the instances and the previous default value of"},{"line_number":8,"context_line":"    \"\u003cnumber of CPUs\u003e / 2\" did not make sense for ML2/OVN as the ovn"},{"line_number":9,"context_line":"    metadata agents are distributed running on Compute nodes instead of"},{"line_number":10,"context_line":"    Controller nodes. In fact, the previous default value could cause"},{"line_number":11,"context_line":"    scalability issues with ML2/OVN and was overwritten by the deployment"}],"source_content_type":"text/x-yaml","patch_set":3,"id":"9f560f44_d5e2e6d3","line":8,"range":{"start_line":8,"start_character":65,"end_line":8,"end_character":68},"in_reply_to":"9f560f44_7d22eba4","updated":"2020-09-09 08:14:55.000000000","message":"Done","commit_id":"b5804a0b1b2bdca03f64d74ff9aac5e6c5cfcb7d"}]}