)]}'
{"specs/train/undercloud-minion.rst":[{"author":{"_account_id":24245,"name":"Harald Jensås","email":"hjensas@redhat.com","username":"harald.jensas"},"change_message_id":"adb0d1ba8cea9dd0ce33791e6014d82250d05bb8","unresolved":false,"context_lines":[{"line_number":16,"context_line":"when creating and updating stacks for deployment.  By adding a new light"},{"line_number":17,"context_line":"weight minion node, we can scale the Heat capacity horizontally."},{"line_number":18,"context_line":""},{"line_number":19,"context_line":"Additionally since these nodes could be more remote, we could add an"},{"line_number":20,"context_line":"ironic-conductor instance to be able to manage hosts in a remote region"},{"line_number":21,"context_line":"while still having a central undercloud for the main management."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"bfb3d3c7_993b3c9b","line":20,"range":{"start_line":19,"start_character":0,"end_line":20,"end_character":71},"updated":"2019-05-17 13:22:56.000000000","message":"It would be good for L3 networking if we could also have the option to place a neutron dhcp agent on the minion node. This would replace the use of external DHCP relay service that we currently need to forward DHCP requests back to the undercloud.","commit_id":"c12cd26c216e8ba657b89385e00edcd789e8a576"},{"author":{"_account_id":6926,"name":"Bogdan Dobrelya","email":"bdobreli@redhat.com","username":"bogdando"},"change_message_id":"6215cfcfc30b60d14fc7c2addcd35077a7ebdb99","unresolved":false,"context_lines":[{"line_number":78,"context_line":"  resource capacity being provided."},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"* Locating an ironic-conductor closer to the nodes being managed can improve"},{"line_number":81,"context_line":"  performance by being closer to the systems (less latency, etc)."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Other Deployer Impact"}],"source_content_type":"text/x-rst","patch_set":1,"id":"dfbec78f_39054dc3","line":81,"updated":"2019-05-06 15:41:14.000000000","message":"I wonder if that also removes L2 over WAN requirements for provisioning at the edge?","commit_id":"c12cd26c216e8ba657b89385e00edcd789e8a576"},{"author":{"_account_id":14985,"name":"Alex Schultz","email":"aschultz@next-development.com","username":"mwhahaha"},"change_message_id":"9e791a7043121f1f42e66e3c4b1402fac06559ff","unresolved":false,"context_lines":[{"line_number":78,"context_line":"  resource capacity being provided."},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"* Locating an ironic-conductor closer to the nodes being managed can improve"},{"line_number":81,"context_line":"  performance by being closer to the systems (less latency, etc)."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Other Deployer Impact"}],"source_content_type":"text/x-rst","patch_set":1,"id":"dfbec78f_69516662","line":81,"in_reply_to":"dfbec78f_1d926c9f","updated":"2019-05-10 14:00:11.000000000","message":"It\u0027s a technical detail as to how ironic-conductor is configured on a minion. This spec is primarily to establish a tool to allow us to deploy a minion host.  The main focus is on heat-engine, ironic-conductor comes secondary and we\u0027ll have to investigate the specifics as to what it means to deploy an ironic-conductor on the edge. At the moment this is primarily for scaling the heat processing capacity.","commit_id":"c12cd26c216e8ba657b89385e00edcd789e8a576"},{"author":{"_account_id":10239,"name":"Dmitry Tantsur","email":"dtantsur@protonmail.com","username":"dtantsur"},"change_message_id":"d99d3334d02e9d97ff135e7ad82382f418aa788e","unresolved":false,"context_lines":[{"line_number":78,"context_line":"  resource capacity being provided."},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"* Locating an ironic-conductor closer to the nodes being managed can improve"},{"line_number":81,"context_line":"  performance by being closer to the systems (less latency, etc)."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Other Deployer Impact"}],"source_content_type":"text/x-rst","patch_set":1,"id":"dfbec78f_44a81176","line":81,"in_reply_to":"dfbec78f_39054dc3","updated":"2019-05-10 07:32:35.000000000","message":"It does. This spec misses configuring conductor groups to establish node-to-conductor mapping, but I guess it\u0027s a technical detail. We may switch ironic from rabbitmq to JSON RPC in Edge case to avoid stretching rabbitmq over WAN, however, I don\u0027t think it would help heat-engine.","commit_id":"c12cd26c216e8ba657b89385e00edcd789e8a576"},{"author":{"_account_id":6926,"name":"Bogdan Dobrelya","email":"bdobreli@redhat.com","username":"bogdando"},"change_message_id":"3666bc7d13936295561e6496495a17ef750fc7be","unresolved":false,"context_lines":[{"line_number":78,"context_line":"  resource capacity being provided."},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"* Locating an ironic-conductor closer to the nodes being managed can improve"},{"line_number":81,"context_line":"  performance by being closer to the systems (less latency, etc)."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":""},{"line_number":84,"context_line":"Other Deployer Impact"}],"source_content_type":"text/x-rst","patch_set":1,"id":"dfbec78f_1d926c9f","line":81,"in_reply_to":"dfbec78f_44a81176","updated":"2019-05-10 09:56:02.000000000","message":"For Edge deployments, like DCN where you have to provision HW over WANs mostly w/o L2 (and/or IPv4 at all) that is not just a technical detail, but a KEY FEATURE that worths being noticed here with and caps locked, really","commit_id":"c12cd26c216e8ba657b89385e00edcd789e8a576"}]}