)]}'
{"specs/kilo/approved/qualified-uuids.rst":[{"author":{"_account_id":100,"name":"Johannes Erdfelt","email":"johannes@erdfelt.com","username":"johannes.erdfelt"},"change_message_id":"b837b6930ee1e1f6da992ee1a65f6c0cab627662","unresolved":false,"context_lines":[{"line_number":52,"context_line":"                         endpoint_type\u003dendpoint_type)"},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"This will need to change in order to support multiple providers of"},{"line_number":55,"context_line":"cinder block storage services."},{"line_number":56,"context_line":""},{"line_number":57,"context_line":"Please note that there is separate work necessary to facilitate the actual"},{"line_number":58,"context_line":"movement of data: networks for Neutron or blocks for Cinder. This spec is"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_0cbb5776","line":55,"updated":"2014-12-15 17:01:21.000000000","message":"Why can\u0027t this be implemented as a proxy that understands what the backend cinder providers are?\n\nI guess what I\u0027m asking is why fundamentally does this complexity need to be pushed into nova?","commit_id":"1d2bc84bc7d6d75088c9b6ae984b5fcb41db58b3"},{"author":{"_account_id":2218,"name":"Adam Young","email":"adam@younglogic.com","username":"ayoung"},"change_message_id":"d4fb1423524a4865c1ae50bee0766c09ada9108f","unresolved":false,"context_lines":[{"line_number":52,"context_line":"                         endpoint_type\u003dendpoint_type)"},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"This will need to change in order to support multiple providers of"},{"line_number":55,"context_line":"cinder block storage services."},{"line_number":56,"context_line":""},{"line_number":57,"context_line":"Please note that there is separate work necessary to facilitate the actual"},{"line_number":58,"context_line":"movement of data: networks for Neutron or blocks for Cinder. This spec is"}],"source_content_type":"text/x-rst","patch_set":1,"id":"7aaa499b_4dfc853e","line":55,"in_reply_to":"3a961159_0cbb5776","updated":"2015-04-10 18:52:46.000000000","message":"Its not just Cinder;  every remote reference that is managed by a remote server would need the same proxy, and that proxy would need to maintain lookup tables for every value it stored.  Since we made the decision to make Nova, Glance, Cinder, and Neutron separate projects, we should take the logical steps and allow them to clearly be separable. \n\nThere is also the Keystone catalog to consider, which would have to list the endpoint tyothe end user, but then tell the compute node to always use the same node (proxy)  which is a fairly ugly solution.","commit_id":"1d2bc84bc7d6d75088c9b6ae984b5fcb41db58b3"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"0e6bf12ae1d85ec6bd3f3f548b4e57e96d1b9172","unresolved":false,"context_lines":[{"line_number":78,"context_line":"another on the higher quality Provider C. The customer wants 1 VM to be"},{"line_number":79,"context_line":"on the cinder instance maintained by Provider B and the other on"},{"line_number":80,"context_line":"Provider C. Currently, there is no way to identify which of the two"},{"line_number":81,"context_line":"cinder providers a object is coming from given just the objects UUID."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":"Project Priority"},{"line_number":84,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"5a890539_113f6cd7","line":81,"updated":"2014-11-13 20:21:21.000000000","message":"I think we are a ways off from being able to be able to make services from two different providers work seamlessly together. That being said this is an interesting proposal, since this has a larger scope then just nova (in the long run) perhaps this is a good topic to take to the openstack-dev ML.","commit_id":"1d2bc84bc7d6d75088c9b6ae984b5fcb41db58b3"},{"author":{"_account_id":13812,"name":"Jay Hennessey","email":"henn@users.sourceforge.net","username":"henn"},"change_message_id":"b55e79aca1d3a6d8347ed0ff2aa86f43c8929d4f","unresolved":false,"context_lines":[{"line_number":78,"context_line":"another on the higher quality Provider C. The customer wants 1 VM to be"},{"line_number":79,"context_line":"on the cinder instance maintained by Provider B and the other on"},{"line_number":80,"context_line":"Provider C. Currently, there is no way to identify which of the two"},{"line_number":81,"context_line":"cinder providers a object is coming from given just the objects UUID."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":"Project Priority"},{"line_number":84,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"7aaa499b_1e31020c","line":81,"in_reply_to":"5a890539_113f6cd7","updated":"2015-04-11 00:17:07.000000000","message":"Thanks for your comment and suggestion. I agree that there is still more work to do to have true interoperability. This is one piece of a broader (and longer-lasting) effort.","commit_id":"1d2bc84bc7d6d75088c9b6ae984b5fcb41db58b3"},{"author":{"_account_id":2218,"name":"Adam Young","email":"adam@younglogic.com","username":"ayoung"},"change_message_id":"d4fb1423524a4865c1ae50bee0766c09ada9108f","unresolved":false,"context_lines":[{"line_number":115,"context_line":""},{"line_number":116,"context_line":"    045a32e7-7eab-4b87-bc87-d7bd148f4558@https://cinder.provider.com:2600/v2"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"There may be an additional database storage cost associated with this,"},{"line_number":119,"context_line":"as IDs are currently stored as varchar(36)."},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Though minimal, several pieces of code at the intersections of services that"}],"source_content_type":"text/x-rst","patch_set":1,"id":"7aaa499b_8d5e0d44","line":118,"updated":"2015-04-10 18:52:46.000000000","message":"If there is an index issue, the value can be split in the database betwen the volume identifier and the endpoint, and stored in separate columns.  Then the set of endpoints could be referenced by id, and both values should be able to fit into one index-able column.  Something like:\n\n\"[\u0027045a32e7-7eab-4b87-bc87-d7bd148f4558\u0027,\u002712\u0027]\"\n\nWhere the \"12\" is the index into the endpoints table, \n12:   https://cinder.provider.com:2600/v2/\n\nand used for expanding the string ID above back to \n https://cinder.provider.com:2600/v2/045a32e7-7eab-4b87-bc87-d7bd148f4558\n\nHowever, that is a performance tune.  It is far more likely that we can expand the column to varchar 255 and just force everyone to keep their URLS short enough to be indexed.\n\nIf indexing is not an issue, then we should convert the table to a text field.","commit_id":"1d2bc84bc7d6d75088c9b6ae984b5fcb41db58b3"}],"specs/liberty/approved/qualified-uuids.rst":[{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":8,"context_line":"Qualified UUIDs to support multiple endpoints for a particular service type"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/qualified-uuids"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Nova makes the assumption that there is only one endpoint for each type of"},{"line_number":14,"context_line":"service. This blueprint outlines some use cases on why supporting multiple"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_2955b51c","line":11,"updated":"2015-04-29 16:23:56.000000000","message":"really what you mean here is support-multiple-cinder deployments?","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":16355,"name":"Orran Krieger","email":"okrieg@bu.edu","username":"okrieg"},"change_message_id":"9e46093fadacc53c3ca9f6fed33ec88290fc4c7c","unresolved":false,"context_lines":[{"line_number":8,"context_line":"Qualified UUIDs to support multiple endpoints for a particular service type"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/qualified-uuids"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Nova makes the assumption that there is only one endpoint for each type of"},{"line_number":14,"context_line":"service. This blueprint outlines some use cases on why supporting multiple"}],"source_content_type":"text/x-rst","patch_set":3,"id":"9a0c5dc1_d9ab4488","line":11,"in_reply_to":"3acd31a7_2955b51c","updated":"2015-05-14 17:54:27.000000000","message":"Eventually we plan on having multiple co-existing deployments of most services, including cinder, glance, neutron, ..., but cinder is definitely the first priority.  Newbie question, is it better to have one blueprint for all, or just focus on the first service that this applies to?","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":23,"context_line":"the OpenStack community, requests feedback on the best direction for realizing"},{"line_number":24,"context_line":"our goal of mixing and matching services under Nova. The current proposal"},{"line_number":25,"context_line":"focuses on cinder, though we suspect that Nova/Neutron interactions are"},{"line_number":26,"context_line":"similarly scoped."},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Problem description"},{"line_number":29,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_898969cd","line":26,"updated":"2015-04-29 16:23:56.000000000","message":"I think we need a summit session on this, could you drop my an email to discuss how/where we should fit that in, and if you are able to attend.","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":16355,"name":"Orran Krieger","email":"okrieg@bu.edu","username":"okrieg"},"change_message_id":"9e46093fadacc53c3ca9f6fed33ec88290fc4c7c","unresolved":false,"context_lines":[{"line_number":23,"context_line":"the OpenStack community, requests feedback on the best direction for realizing"},{"line_number":24,"context_line":"our goal of mixing and matching services under Nova. The current proposal"},{"line_number":25,"context_line":"focuses on cinder, though we suspect that Nova/Neutron interactions are"},{"line_number":26,"context_line":"similarly scoped."},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Problem description"},{"line_number":29,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"9a0c5dc1_f9b968cc","line":26,"in_reply_to":"3acd31a7_898969cd","updated":"2015-05-14 17:54:27.000000000","message":"email bounced, yes, we will be at summit, would love a session on this, sent you email","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":78,"context_line":"another on the higher quality Provider C. The customer wants 1 VM to be"},{"line_number":79,"context_line":"on the cinder instance maintained by Provider B and the other on"},{"line_number":80,"context_line":"Provider C. Currently, there is no way to identify which of the two"},{"line_number":81,"context_line":"cinder providers a object is coming from given just the objects UUID."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":"Project Priority"},{"line_number":84,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_bb2468af","line":81,"updated":"2015-04-29 16:23:56.000000000","message":"This all assumes a fully shared keystone deployment I assume? or is there a dependency of keystone federation of some kind?","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":16355,"name":"Orran Krieger","email":"okrieg@bu.edu","username":"okrieg"},"change_message_id":"9e46093fadacc53c3ca9f6fed33ec88290fc4c7c","unresolved":false,"context_lines":[{"line_number":78,"context_line":"another on the higher quality Provider C. The customer wants 1 VM to be"},{"line_number":79,"context_line":"on the cinder instance maintained by Provider B and the other on"},{"line_number":80,"context_line":"Provider C. Currently, there is no way to identify which of the two"},{"line_number":81,"context_line":"cinder providers a object is coming from given just the objects UUID."},{"line_number":82,"context_line":""},{"line_number":83,"context_line":"Project Priority"},{"line_number":84,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"7a016987_30d62cc6","line":81,"in_reply_to":"3acd31a7_bb2468af","updated":"2015-05-14 17:54:27.000000000","message":"Yes, we are working on keystone federation necessary for this","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":113,"context_line":""},{"line_number":114,"context_line":"::"},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"    045a32e7-7eab-4b87-bc87-d7bd148f4558@https://cinder.provider.com:2600/v2"},{"line_number":117,"context_line":""},{"line_number":118,"context_line":"There may be an additional database storage cost associated with this,"},{"line_number":119,"context_line":"as IDs are currently stored as varchar(36)."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_9baf6ce3","line":116,"updated":"2015-04-29 16:23:56.000000000","message":"honestly, I think you would want a \"cinder endpoint\" uuid, or something. URLs have a habit of changing.\n\nBut it does feel like a lot of extra complexity. And why not just add a new separate field for the provider, that would be much neater than parsing strings.","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":126,"context_line":""},{"line_number":127,"context_line":"This has the advantage of not requiring cinder or neutron changes to start, as"},{"line_number":128,"context_line":"the provider portion of the UUID could be stripped before sending to the remote"},{"line_number":129,"context_line":"service."},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"Other ID modifications"},{"line_number":132,"context_line":"^^^^^^^^^^^^^^^^^^^^^^"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_7b9370a4","line":129,"updated":"2015-04-29 16:23:56.000000000","message":"You are only hitting the tip of the iceburg here.\n\nThis would change all the public APIs to make you specify a bigger uuid, or a provider config :(","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":143,"context_line":"    volume;https://cinder.provider.com:2600/v2/BFoy536rS4e8h9e9FI9FWA\u003d\u003d"},{"line_number":144,"context_line":""},{"line_number":145,"context_line":"Alternatives"},{"line_number":146,"context_line":"------------"},{"line_number":147,"context_line":"Separate provider string"},{"line_number":148,"context_line":"^^^^^^^^^^^^^^^^^^^^^^^^"},{"line_number":149,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_3b5298de","line":146,"updated":"2015-04-29 16:23:56.000000000","message":"Please add the great alternative johannes pointed out, having a proxy layer above neutron/cinder to aggregate multiple neutron and cinder deployments.\n\nThat would seem much simpler here, and is more likely to work around issues with authentication, etc.","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":16355,"name":"Orran Krieger","email":"okrieg@bu.edu","username":"okrieg"},"change_message_id":"9e46093fadacc53c3ca9f6fed33ec88290fc4c7c","unresolved":false,"context_lines":[{"line_number":143,"context_line":"    volume;https://cinder.provider.com:2600/v2/BFoy536rS4e8h9e9FI9FWA\u003d\u003d"},{"line_number":144,"context_line":""},{"line_number":145,"context_line":"Alternatives"},{"line_number":146,"context_line":"------------"},{"line_number":147,"context_line":"Separate provider string"},{"line_number":148,"context_line":"^^^^^^^^^^^^^^^^^^^^^^^^"},{"line_number":149,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"7a016987_709a6416","line":146,"in_reply_to":"3acd31a7_3b5298de","updated":"2015-05-14 17:54:27.000000000","message":"working to integrate that now","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":245,"context_line":"if that is ever needed. For example, if the user shut down a VM and later"},{"line_number":246,"context_line":"started it up, would that user need to again pass a token that includes the"},{"line_number":247,"context_line":"cinder endpoint? What about the case where a VM migrates within a Nova"},{"line_number":248,"context_line":"provider, and a new session is needed for the block storage on the new node?"},{"line_number":249,"context_line":""},{"line_number":250,"context_line":"Endpoint restrictions"},{"line_number":251,"context_line":"^^^^^^^^^^^^^^^^^^^^^"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_3b0058aa","line":248,"updated":"2015-04-29 16:23:56.000000000","message":"thats an existing issue, although the cinder connection is current kept when the VM shuts down, so thats not really an issue here.","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e63ecebe6f02c57953d96fcd1ab3d97257cf0d65","unresolved":false,"context_lines":[{"line_number":310,"context_line":"-----------"},{"line_number":311,"context_line":""},{"line_number":312,"context_line":"Primary assignee:"},{"line_number":313,"context_line":"\tTBD"},{"line_number":314,"context_line":""},{"line_number":315,"context_line":"Other contributors:"},{"line_number":316,"context_line":"\tThe Mass Open Cloud consortium"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_4977d102","line":313,"updated":"2015-04-29 16:23:56.000000000","message":"If this is unassigned, please add to the /backlog/ directory, rather than the approved for release directory.\n\nalso, please remove the tabs.","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"},{"author":{"_account_id":16355,"name":"Orran Krieger","email":"okrieg@bu.edu","username":"okrieg"},"change_message_id":"9e46093fadacc53c3ca9f6fed33ec88290fc4c7c","unresolved":false,"context_lines":[{"line_number":310,"context_line":"-----------"},{"line_number":311,"context_line":""},{"line_number":312,"context_line":"Primary assignee:"},{"line_number":313,"context_line":"\tTBD"},{"line_number":314,"context_line":""},{"line_number":315,"context_line":"Other contributors:"},{"line_number":316,"context_line":"\tThe Mass Open Cloud consortium"}],"source_content_type":"text/x-rst","patch_set":3,"id":"7a016987_506d2803","line":313,"in_reply_to":"3acd31a7_4977d102","updated":"2015-05-14 17:54:27.000000000","message":"we are willing to take this on if approved","commit_id":"900c5e759f4b15f92538ee2bee9fae4304dbd936"}]}