)]}'
{"doc/source/devref/index.rst":[{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"ec81c931a9e9dd3bb0f44c2ed6f9311d64be20df","unresolved":false,"context_lines":[{"line_number":57,"context_line":""},{"line_number":58,"context_line":"   kilo.blueprints"},{"line_number":59,"context_line":"   policies"},{"line_number":60,"context_line":"   project_scope"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Testing guides"},{"line_number":63,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_26c8c8a7","line":60,"updated":"2015-05-08 18:39:25.000000000","message":"since we link to this on the top level do we need this here anymore? Maybe?","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"62fce8ab335db5d7b37e0c17629ed62544096fe3","unresolved":false,"context_lines":[{"line_number":57,"context_line":""},{"line_number":58,"context_line":"   kilo.blueprints"},{"line_number":59,"context_line":"   policies"},{"line_number":60,"context_line":"   project_scope"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Testing guides"},{"line_number":63,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":12,"id":"9af37de9_04b41af0","line":60,"in_reply_to":"1aca2d91_26c8c8a7","updated":"2015-05-11 09:15:32.000000000","message":"I went for both, as I couldn\u0027t decide, lets fix this later.","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"}],"doc/source/devref/project_scope.rst":[{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"0a6bd9b8aa04890d2b732833147b3031c6c92706","unresolved":false,"context_lines":[{"line_number":14,"context_line":"Scope of the Nova project"},{"line_number":15,"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"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Nova is focusing on doing an awesome job of its core mission."},{"line_number":18,"context_line":"This document aims to clarify that core mission."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"By defining what Nova should not do, we allow other projects to own"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_d4fe8a6a","line":17,"updated":"2015-05-01 18:16:14.000000000","message":"\u0027awesome job\u0027! haha","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":33,"context_line":""},{"line_number":34,"context_line":"To implement services and associated libraries to provide massively"},{"line_number":35,"context_line":"scalable, on demand, self service access to compute resources, including"},{"line_number":36,"context_line":"bare metal, virtual machines, and containers."},{"line_number":37,"context_line":""},{"line_number":38,"context_line":"Containers"},{"line_number":39,"context_line":"----------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_1271b9e0","line":36,"range":{"start_line":36,"start_character":34,"end_line":36,"end_character":44},"updated":"2015-05-01 14:56:20.000000000","message":"I kinda wonder if we should remove this from our official mission at this point, but that\u0027s a different discussion I guess.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"5724e356d4eb384e545916ff1f8786c88fc2bc2c","unresolved":false,"context_lines":[{"line_number":33,"context_line":""},{"line_number":34,"context_line":"To implement services and associated libraries to provide massively"},{"line_number":35,"context_line":"scalable, on demand, self service access to compute resources, including"},{"line_number":36,"context_line":"bare metal, virtual machines, and containers."},{"line_number":37,"context_line":""},{"line_number":38,"context_line":"Containers"},{"line_number":39,"context_line":"----------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_8847dc4a","line":36,"range":{"start_line":36,"start_character":34,"end_line":36,"end_character":44},"in_reply_to":"3acd31a7_1271b9e0","updated":"2015-05-01 15:19:17.000000000","message":"its tempting, yeah.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":43,"context_line":"maintain the integrity and usefulness of the existing Nova API."},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"However we are not designed to spin up new containers for every apache"},{"line_number":46,"context_line":"request, nor do we plan to control what goes on inside a containers."},{"line_number":47,"context_line":"They get the same metadata provided to them as virtual machines, to do"},{"line_number":48,"context_line":"with as they see fit."},{"line_number":49,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_724b4d88","line":46,"range":{"start_line":46,"start_character":55,"end_line":46,"end_character":56},"updated":"2015-05-01 14:56:20.000000000","message":"s/a//","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":64,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":65,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":66,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":67,"context_line":"it is expected that gap with shrink over time."},{"line_number":68,"context_line":""},{"line_number":69,"context_line":"Driver Parity"},{"line_number":70,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_b259e55c","line":67,"updated":"2015-05-01 14:56:20.000000000","message":"TBH, I wish we\u0027d remove baremetal from our mission statement as well, and just say \"compute resources\" for which one example is (currently) driving ironic.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"0a6bd9b8aa04890d2b732833147b3031c6c92706","unresolved":false,"context_lines":[{"line_number":64,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":65,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":66,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":67,"context_line":"it is expected that gap with shrink over time."},{"line_number":68,"context_line":""},{"line_number":69,"context_line":"Driver Parity"},{"line_number":70,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_b712f007","line":67,"in_reply_to":"3acd31a7_286fa8c6","updated":"2015-05-01 18:16:14.000000000","message":"++","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"5724e356d4eb384e545916ff1f8786c88fc2bc2c","unresolved":false,"context_lines":[{"line_number":64,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":65,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":66,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":67,"context_line":"it is expected that gap with shrink over time."},{"line_number":68,"context_line":""},{"line_number":69,"context_line":"Driver Parity"},{"line_number":70,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_286fa8c6","line":67,"in_reply_to":"3acd31a7_b259e55c","updated":"2015-05-01 15:19:17.000000000","message":"yeah, we should fix that up.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"0a6bd9b8aa04890d2b732833147b3031c6c92706","unresolved":false,"context_lines":[{"line_number":71,"context_line":""},{"line_number":72,"context_line":"The key tenant of driver parity is that if a feature is supported in a driver,"},{"line_number":73,"context_line":"it should feel the same to users, as if they where using any of the other"},{"line_number":74,"context_line":"drivers that also support that feature. The exception is that, if possible for"},{"line_number":75,"context_line":"widely different performance characteristics, but the effect of that API call"},{"line_number":76,"context_line":"should be identical."},{"line_number":77,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_57f27cd3","line":74,"updated":"2015-05-01 18:16:14.000000000","message":"another way to say this is:\n\nUsers should never need to know or care about which driver a cloud uses. Or even what the word \u0027driver\u0027 means.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"0a6bd9b8aa04890d2b732833147b3031c6c92706","unresolved":false,"context_lines":[{"line_number":90,"context_line":"on demand compute resources, rather than expose all features of all"},{"line_number":91,"context_line":"hypervisors. Where the details of the underlying hypervisor leak through"},{"line_number":92,"context_line":"our APIs, we have failed in this goal, and we should work towards better"},{"line_number":93,"context_line":"abstractions that are more interoperable."},{"line_number":94,"context_line":""},{"line_number":95,"context_line":"Deployment and Packaging"},{"line_number":96,"context_line":"-------------------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_77ca1870","line":93,"updated":"2015-05-01 18:16:14.000000000","message":"I was wondering when you would pull at the \u0027interoperable\u0027 buzzword.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"0a6bd9b8aa04890d2b732833147b3031c6c92706","unresolved":false,"context_lines":[{"line_number":111,"context_line":"Our mission includes the text \"massively scalable\". Lets discuss what that"},{"line_number":112,"context_line":"means."},{"line_number":113,"context_line":""},{"line_number":114,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":115,"context_line":"servers and compute node. Its possible to have a small cloud with lots of"},{"line_number":116,"context_line":"requests for very short lived VMs, or a large cloud with lots of longer"},{"line_number":117,"context_line":"lived VMs. These need to scale out different components of the Nova system"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_977db41e","line":114,"updated":"2015-05-01 18:16:14.000000000","message":"I count 3\n* API requests\n* Physical boxes\n* Servers/Instances","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":112,"context_line":"means."},{"line_number":113,"context_line":""},{"line_number":114,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":115,"context_line":"servers and compute node. Its possible to have a small cloud with lots of"},{"line_number":116,"context_line":"requests for very short lived VMs, or a large cloud with lots of longer"},{"line_number":117,"context_line":"lived VMs. These need to scale out different components of the Nova system"},{"line_number":118,"context_line":"to reach their required level of scale."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_0d0d1e37","line":115,"range":{"start_line":115,"start_character":20,"end_line":115,"end_character":24},"updated":"2015-05-01 14:56:20.000000000","message":"s/node/nodes/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":117,"context_line":"lived VMs. These need to scale out different components of the Nova system"},{"line_number":118,"context_line":"to reach their required level of scale."},{"line_number":119,"context_line":""},{"line_number":120,"context_line":"Ideally all Nova component are either scaled out to match the number of API"},{"line_number":121,"context_line":"requests and builds requests, or scaled out to match the number of running"},{"line_number":122,"context_line":"servers. If we create components that have their load increased relative to"},{"line_number":123,"context_line":"both of these items, we can run into inefficients or resource contention."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_cd0f262d","line":120,"range":{"start_line":120,"start_character":17,"end_line":120,"end_character":26},"updated":"2015-05-01 14:56:20.000000000","message":"s/component/components/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":118,"context_line":"to reach their required level of scale."},{"line_number":119,"context_line":""},{"line_number":120,"context_line":"Ideally all Nova component are either scaled out to match the number of API"},{"line_number":121,"context_line":"requests and builds requests, or scaled out to match the number of running"},{"line_number":122,"context_line":"servers. If we create components that have their load increased relative to"},{"line_number":123,"context_line":"both of these items, we can run into inefficients or resource contention."},{"line_number":124,"context_line":"Although it is possible to make that work in some cases, this should always"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_8d192e7c","line":121,"range":{"start_line":121,"start_character":13,"end_line":121,"end_character":19},"updated":"2015-05-01 14:56:20.000000000","message":"s/builds/build/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":120,"context_line":"Ideally all Nova component are either scaled out to match the number of API"},{"line_number":121,"context_line":"requests and builds requests, or scaled out to match the number of running"},{"line_number":122,"context_line":"servers. If we create components that have their load increased relative to"},{"line_number":123,"context_line":"both of these items, we can run into inefficients or resource contention."},{"line_number":124,"context_line":"Although it is possible to make that work in some cases, this should always"},{"line_number":125,"context_line":"be considered."},{"line_number":126,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_4d2af67f","line":123,"range":{"start_line":123,"start_character":37,"end_line":123,"end_character":49},"updated":"2015-05-01 14:56:20.000000000","message":"inefficiencies?","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":138,"context_line":"----------------------------"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"Nova is widely used in production. As such we need to respect the needs of our"},{"line_number":141,"context_line":"existing users."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"As a community we are aiming towards a smooth upgrade process, where users"},{"line_number":144,"context_line":"should be unaware you have just upgraded your deployment, except that there"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_4d51d60d","line":141,"updated":"2015-05-01 14:56:20.000000000","message":"Can we say something here about CDers and making sure it works for people consuming master and stable?","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"5724e356d4eb384e545916ff1f8786c88fc2bc2c","unresolved":false,"context_lines":[{"line_number":138,"context_line":"----------------------------"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"Nova is widely used in production. As such we need to respect the needs of our"},{"line_number":141,"context_line":"existing users."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"As a community we are aiming towards a smooth upgrade process, where users"},{"line_number":144,"context_line":"should be unaware you have just upgraded your deployment, except that there"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_e869b0bf","line":141,"in_reply_to":"3acd31a7_4d51d60d","updated":"2015-05-01 15:19:17.000000000","message":"very good idea, will add that.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":145,"context_line":"might be additional feature available and improved stability and performance"},{"line_number":146,"context_line":"of some existing features."},{"line_number":147,"context_line":""},{"line_number":148,"context_line":"We don\u0027t ever want to remove features are users rely on. Sometimes we need to"},{"line_number":149,"context_line":"migrate users to a new implementation of that feature, which may require extra"},{"line_number":150,"context_line":"steps by the deployer, but the end users should be unaffected by such changes."},{"line_number":151,"context_line":"However there are times when some features become a problem to maintain, and"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_0da99edf","line":148,"range":{"start_line":148,"start_character":38,"end_line":148,"end_character":41},"updated":"2015-05-01 14:56:20.000000000","message":"s/are/our/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":154,"context_line":"Ideally we are able to rework the feature so it can be maintained, but in some"},{"line_number":155,"context_line":"rare cases, the feature no longer works, is not tested, and no one is stepping"},{"line_number":156,"context_line":"forward to maintain that feature, the best option can be to remove that"},{"line_number":157,"context_line":"feature."},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"When we remove features, we need warn users by first marking those features as"},{"line_number":160,"context_line":"deprecated, before we finally remove the feature. The idea is to get feedback"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_ed97828e","line":157,"updated":"2015-05-01 14:56:20.000000000","message":"+1","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":179,"context_line":"and if you use both Nova API and Neutron API, the miss match can lead to"},{"line_number":180,"context_line":"some very unexpected results, in certain cases."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Our intention is to avoid add to the problems we already have in this area."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":"No more Orchestration"},{"line_number":185,"context_line":"----------------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_0df75ef6","line":182,"range":{"start_line":182,"start_character":26,"end_line":182,"end_character":29},"updated":"2015-05-01 14:56:20.000000000","message":"s/add/adding/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":200,"context_line":"rather than getting distracted with the orchestration problem discussed above."},{"line_number":201,"context_line":""},{"line_number":202,"context_line":"Secondly, there are many ways to monitor the health of your infrastructure."},{"line_number":203,"context_line":"Nova does not intent to try and replicate their great work, instead we look"},{"line_number":204,"context_line":"to ensure we work well with the existing monitoring ecosystem."},{"line_number":205,"context_line":""},{"line_number":206,"context_line":"No native support for Batch Processing"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_2dcb3aa7","line":203,"range":{"start_line":203,"start_character":14,"end_line":203,"end_character":20},"updated":"2015-05-01 14:56:20.000000000","message":"s/intent/indend/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":206,"context_line":"No native support for Batch Processing"},{"line_number":207,"context_line":"---------------------------------------"},{"line_number":208,"context_line":""},{"line_number":209,"context_line":"Currently Nova focuses on providing on demand compute resources in the style"},{"line_number":210,"context_line":"of classic Infrastructure as a Service clouds. A large pool of compute"},{"line_number":211,"context_line":"resources that people can consume in a self-service way."},{"line_number":212,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_cdc0467f","line":209,"range":{"start_line":209,"start_character":36,"end_line":209,"end_character":45},"updated":"2015-05-01 14:56:20.000000000","message":"Should be \"on-demand\" I think","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":207,"context_line":"---------------------------------------"},{"line_number":208,"context_line":""},{"line_number":209,"context_line":"Currently Nova focuses on providing on demand compute resources in the style"},{"line_number":210,"context_line":"of classic Infrastructure as a Service clouds. A large pool of compute"},{"line_number":211,"context_line":"resources that people can consume in a self-service way."},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_8dba4e09","line":210,"range":{"start_line":210,"start_character":11,"end_line":210,"end_character":38},"updated":"2015-05-01 14:56:20.000000000","message":"Should be Infrastructure-as-a-service I think.","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":211,"context_line":"resources that people can consume in a self-service way."},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"},{"line_number":214,"context_line":"for compute resources compared with the above of compute resource thats"},{"line_number":215,"context_line":"currently available. We generally assume a level of spare capacity is"},{"line_number":216,"context_line":"maintained for future requests. As such our quota system is more"},{"line_number":217,"context_line":"focused on limiting the current level of resource usage, rather than"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_8d512edb","line":214,"range":{"start_line":214,"start_character":40,"end_line":214,"end_character":48},"updated":"2015-05-01 14:56:20.000000000","message":"Not sure what to suggest here, but I think this was unintentional","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":219,"context_line":""},{"line_number":220,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":221,"context_line":"batch job processing."},{"line_number":222,"context_line":"But the current focus is how to layer such an abstraction on top of the basic"},{"line_number":223,"context_line":"primitives Nova currently provides, possibly adding additional API where that"},{"line_number":224,"context_line":"makes good sense. Should this turn out to be impractical, we may have to"},{"line_number":225,"context_line":"revise our approach."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_8d680e89","line":222,"range":{"start_line":222,"start_character":22,"end_line":222,"end_character":24},"updated":"2015-05-01 14:56:20.000000000","message":"s/is/is on/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":220,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":221,"context_line":"batch job processing."},{"line_number":222,"context_line":"But the current focus is how to layer such an abstraction on top of the basic"},{"line_number":223,"context_line":"primitives Nova currently provides, possibly adding additional API where that"},{"line_number":224,"context_line":"makes good sense. Should this turn out to be impractical, we may have to"},{"line_number":225,"context_line":"revise our approach."},{"line_number":226,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_2d10da00","line":223,"range":{"start_line":223,"start_character":63,"end_line":223,"end_character":66},"updated":"2015-05-01 14:56:20.000000000","message":"maybe \"primitives\" again or \"APIs\" ?","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"1b992305b5db44d882216503f6e64ef4da6e94b1","unresolved":false,"context_lines":[{"line_number":237,"context_line":"provide third party compatibility on top of the Nova API. Where needed, we"},{"line_number":238,"context_line":"will work this those projects to extending the Nova API such that its"},{"line_number":239,"context_line":"possible to add that functionality on top of the Nova API. However, we do"},{"line_number":240,"context_line":"not indent to add API calls for those services to persist third party API"},{"line_number":241,"context_line":"specific information in the Nova database. Instead we want to focus on"},{"line_number":242,"context_line":"additions that enhance the existing Nova API."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3acd31a7_4d30b69e","line":240,"range":{"start_line":240,"start_character":4,"end_line":240,"end_character":10},"updated":"2015-05-01 14:56:20.000000000","message":"s/indent/intend/","commit_id":"ce976c06aaf5cc3ef8c748a12e902081ae3d76a3"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":17,"context_line":"Nova is focusing on doing an awesome job of its core mission."},{"line_number":18,"context_line":"This document aims to clarify that core mission."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"By defining what Nova should not do, we allow other projects to own"},{"line_number":21,"context_line":"solving problems that are outside our core mission. We aim to work with"},{"line_number":22,"context_line":"those projects that can make use of the existing abstraction Nova provides,"},{"line_number":23,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_2fe3d6a5","line":20,"updated":"2015-05-05 21:28:33.000000000","message":"I think this document should define what Nova should do.  It\u0027s important to also state what Nova doesn\u0027t do, but the focus should be on what Nova is intending to provide.","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"4065ec08769e49d6a8fd82c29aaafdce7ed24944","unresolved":false,"context_lines":[{"line_number":17,"context_line":"Nova is focusing on doing an awesome job of its core mission."},{"line_number":18,"context_line":"This document aims to clarify that core mission."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"By defining what Nova should not do, we allow other projects to own"},{"line_number":21,"context_line":"solving problems that are outside our core mission. We aim to work with"},{"line_number":22,"context_line":"those projects that can make use of the existing abstraction Nova provides,"},{"line_number":23,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"1aca2d91_32bbd910","line":20,"in_reply_to":"3acd31a7_2fe3d6a5","updated":"2015-05-06 08:28:49.000000000","message":"yeah, sorry, good point.\nI am half way through change the direction here, I will update that.","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":24,"context_line":"This is a living document to help record where we agree about what Nova"},{"line_number":25,"context_line":"should not be doing, and why. Please treat this as a discussion of"},{"line_number":26,"context_line":"interesting, and hopefully useful, examples. It is not intended to be a"},{"line_number":27,"context_line":"string and exhaustive policy statement."},{"line_number":28,"context_line":""},{"line_number":29,"context_line":"Our Mission"},{"line_number":30,"context_line":"-----------"}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_efef1e85","line":27,"range":{"start_line":27,"start_character":0,"end_line":27,"end_character":6},"updated":"2015-05-05 21:28:33.000000000","message":"strong","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":35,"context_line":"scalable, on demand, self service access to compute resources"},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"Our official mission statement also includes the following examples of"},{"line_number":38,"context_line":"compute resources: bare metal, virtual machines, and containers."},{"line_number":39,"context_line":""},{"line_number":40,"context_line":"This document should help clarify what the mission statement currently means"},{"line_number":41,"context_line":"to the Nova community."}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_8fa1ca48","line":38,"range":{"start_line":38,"start_character":53,"end_line":38,"end_character":63},"updated":"2015-05-05 21:28:33.000000000","message":"Should it include containers?  Also, where is the mission statement officially at?","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"4065ec08769e49d6a8fd82c29aaafdce7ed24944","unresolved":false,"context_lines":[{"line_number":35,"context_line":"scalable, on demand, self service access to compute resources"},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"Our official mission statement also includes the following examples of"},{"line_number":38,"context_line":"compute resources: bare metal, virtual machines, and containers."},{"line_number":39,"context_line":""},{"line_number":40,"context_line":"This document should help clarify what the mission statement currently means"},{"line_number":41,"context_line":"to the Nova community."}],"source_content_type":"text/x-rst","patch_set":6,"id":"1aca2d91_b213690d","line":38,"range":{"start_line":38,"start_character":53,"end_line":38,"end_character":63},"in_reply_to":"3acd31a7_8fa1ca48","updated":"2015-05-06 08:28:49.000000000","message":"yeah, thats come up quite a lot, I think it need to drop the examples really.","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":52,"context_line":"They get the same metadata provided to them as virtual machines, to do"},{"line_number":53,"context_line":"with as they see fit."},{"line_number":54,"context_line":""},{"line_number":55,"context_line":"Please note the Magnum project has the following mission:"},{"line_number":56,"context_line":"To provide a set of services for management of application containers"},{"line_number":57,"context_line":"in a multitenant cloud environment."},{"line_number":58,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_afb36699","line":55,"updated":"2015-05-05 21:28:33.000000000","message":"Might be good to say something like \"The Nova API is not a good fit for a lot of container use cases.  The Magnum project intends to deliver a good container experience built on top of Nova.\"","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":56,"context_line":"To provide a set of services for management of application containers"},{"line_number":57,"context_line":"in a multitenant cloud environment."},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"Bare Metal Servers"},{"line_number":60,"context_line":"-------------------"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Ironic project has been pioneering the idea of treating physical machines in"}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_ef4b9e85","line":59,"updated":"2015-05-05 21:28:33.000000000","message":"I wonder if we need to call out baremetal servers?  I think it\u0027s important to call out that Nova is intended to manage the lifecycle of things that are, or act like, bootable machines.  And both VMs and baremetal servers are manageable this way.","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"4065ec08769e49d6a8fd82c29aaafdce7ed24944","unresolved":false,"context_lines":[{"line_number":56,"context_line":"To provide a set of services for management of application containers"},{"line_number":57,"context_line":"in a multitenant cloud environment."},{"line_number":58,"context_line":""},{"line_number":59,"context_line":"Bare Metal Servers"},{"line_number":60,"context_line":"-------------------"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Ironic project has been pioneering the idea of treating physical machines in"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1aca2d91_f2a561b5","line":59,"in_reply_to":"3acd31a7_ef4b9e85","updated":"2015-05-06 08:28:49.000000000","message":"its more for completeness, and so its clear we use Ironic, and not have our own baremetal driver","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":220,"context_line":""},{"line_number":221,"context_line":"Our intention is to avoid adding to the problems we already have in this area."},{"line_number":222,"context_line":""},{"line_number":223,"context_line":"No more Orchestration"},{"line_number":224,"context_line":"----------------------"},{"line_number":225,"context_line":""},{"line_number":226,"context_line":"A frequently raised topic is adding a Nova API to move all instances from"}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_129a7d6e","line":223,"updated":"2015-05-05 21:28:33.000000000","message":"In my mind Nova is a low level infrastructure API.  It is plumbing upon which richer ideas should be built.","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"4065ec08769e49d6a8fd82c29aaafdce7ed24944","unresolved":false,"context_lines":[{"line_number":220,"context_line":""},{"line_number":221,"context_line":"Our intention is to avoid adding to the problems we already have in this area."},{"line_number":222,"context_line":""},{"line_number":223,"context_line":"No more Orchestration"},{"line_number":224,"context_line":"----------------------"},{"line_number":225,"context_line":""},{"line_number":226,"context_line":"A frequently raised topic is adding a Nova API to move all instances from"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1aca2d91_5276553f","line":223,"in_reply_to":"3acd31a7_129a7d6e","updated":"2015-05-06 08:28:49.000000000","message":"Thats a good way of putting it. thanks, will add that.\n\nI guess I am try to express the API has some orchestration in it, and we have to keep that now, but we will not add to it.","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"7024ca486b19c55e793e69e5e1fab9cace80d989","unresolved":false,"context_lines":[{"line_number":276,"context_line":"provide third party compatibility on top of the Nova API. Where needed, we"},{"line_number":277,"context_line":"will work this those projects to extending the Nova API such that its"},{"line_number":278,"context_line":"possible to add that functionality on top of the Nova API. However, we do"},{"line_number":279,"context_line":"not indend to add API calls for those services to persist third party API"},{"line_number":280,"context_line":"specific information in the Nova database. Instead we want to focus on"},{"line_number":281,"context_line":"additions that enhance the existing Nova API."}],"source_content_type":"text/x-rst","patch_set":6,"id":"3acd31a7_727e511a","line":279,"range":{"start_line":279,"start_character":4,"end_line":279,"end_character":10},"updated":"2015-05-05 21:28:33.000000000","message":"intend","commit_id":"e374fca27248dd2b66cb87bd2ea3ee475daa2308"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":11,"context_line":"      License for the specific language governing permissions and limitations"},{"line_number":12,"context_line":"      under the License."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"Scope of the Nova project"},{"line_number":15,"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"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Nova is focusing on doing an awesome job of its core mission."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_00698bbe","line":14,"updated":"2015-05-06 19:10:06.000000000","message":"Does this belong in devref or directly in doc/source/\n\nI think it would be nice to find a link to this document directly from http://docs.openstack.org/developer/nova/","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":11,"context_line":"      License for the specific language governing permissions and limitations"},{"line_number":12,"context_line":"      under the License."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"Scope of the Nova project"},{"line_number":15,"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"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Nova is focusing on doing an awesome job of its core mission."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_2b38d6d2","line":14,"in_reply_to":"1aca2d91_00698bbe","updated":"2015-05-07 07:28:53.000000000","message":"yeah, I want to reorganise that, agree it goes there in the end.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":37,"context_line":"http://governance.openstack.org/reference/projects/nova.html#mission"},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"This document should help clarify what the mission statement currently means"},{"line_number":40,"context_line":"to the Nova community."},{"line_number":41,"context_line":""},{"line_number":42,"context_line":"Containers"},{"line_number":43,"context_line":"----------"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_60eeff23","line":40,"updated":"2015-05-06 19:10:06.000000000","message":"The formatting of this section isn\u0027t very nice. I think we can do better\n\nhttp://docs-draft.openstack.org/23/178623/8/check/gate-nova-docs/614c694//doc/build/html/devref/project_scope.html#our-mission","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":38,"context_line":""},{"line_number":39,"context_line":"This document should help clarify what the mission statement currently means"},{"line_number":40,"context_line":"to the Nova community."},{"line_number":41,"context_line":""},{"line_number":42,"context_line":"Containers"},{"line_number":43,"context_line":"----------"},{"line_number":44,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_00218b4d","line":41,"updated":"2015-05-06 19:10:06.000000000","message":"Instead of just having a bunch of random sections below, it would be nice to have a section outlining and breaking down the topics.\n\nSomething like: Design Tenants, a brief list of what Nova is and is not.\n\nDesign Tenants would be:\n\n * Driver parity\n * Scalability\n * Upgradeability \n * Don\u0027t break the user \n\netc.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"40222ba4c14209ba13aac6c1db4011089f52c13b","unresolved":false,"context_lines":[{"line_number":38,"context_line":""},{"line_number":39,"context_line":"This document should help clarify what the mission statement currently means"},{"line_number":40,"context_line":"to the Nova community."},{"line_number":41,"context_line":""},{"line_number":42,"context_line":"Containers"},{"line_number":43,"context_line":"----------"},{"line_number":44,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_d7b6a13c","line":41,"in_reply_to":"1aca2d91_00218b4d","updated":"2015-05-07 09:23:10.000000000","message":"I have attempted grouping the sections instead, in the help that will make it clearer but also be easy to maintain.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":43,"context_line":"----------"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"The Nova API is not a good fit for a lot of container use cases."},{"line_number":46,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":47,"context_line":"on top of Nova."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The Magnum project has the following mission:"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_40dd4341","line":46,"updated":"2015-05-06 19:10:06.000000000","message":"I don\u0027t think we need to explicitly point to another project that does this. We shouldn\u0027t need to say magnum is the way to do this.  If another project becomes in vogue we have to revise this.\n\nI think we can simply say:\n\nNova provides server like instances, that can include containers that look like a full server. But the numerous non server-like container use cases are outside the scope of nova.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":43,"context_line":"----------"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"The Nova API is not a good fit for a lot of container use cases."},{"line_number":46,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":47,"context_line":"on top of Nova."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The Magnum project has the following mission:"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_0944bd18","line":46,"in_reply_to":"1aca2d91_065a6c2a","updated":"2015-05-08 19:52:51.000000000","message":"Yeah, it\u0027s very \"Big Tent\" to not mention anything by name.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"85469dfb4767f23ee44ff1213aba12492d673581","unresolved":false,"context_lines":[{"line_number":43,"context_line":"----------"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"The Nova API is not a good fit for a lot of container use cases."},{"line_number":46,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":47,"context_line":"on top of Nova."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The Magnum project has the following mission:"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_065a6c2a","line":46,"in_reply_to":"1aca2d91_2b3fb6b5","updated":"2015-05-08 18:31:49.000000000","message":"Yes official, but it is not the only way to do things. I don\u0027t think we should \u0027bless\u0027 it.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":43,"context_line":"----------"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"The Nova API is not a good fit for a lot of container use cases."},{"line_number":46,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":47,"context_line":"on top of Nova."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The Magnum project has the following mission:"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_2b3fb6b5","line":46,"in_reply_to":"1aca2d91_40dd4341","updated":"2015-05-07 07:28:53.000000000","message":"Its an official project, so I was trying to give it as an example. I think thats important in this case.\n\nAgreed that adds overhead.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"For example, Nova is not designed to spin up new containers for every apache"},{"line_number":58,"context_line":"request, nor do we plan to control what goes on inside containers."},{"line_number":59,"context_line":"They get the same metadata provided to them as virtual machines, to do"},{"line_number":60,"context_line":"with as they see fit."},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Bare Metal Servers"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_60a6ffc2","line":59,"updated":"2015-05-06 19:10:06.000000000","message":"This is a good point to flesh out in its own section. Nova doesn\u0027t deal with what is inside the \u0027instances\u0027 that it provides.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"For example, Nova is not designed to spin up new containers for every apache"},{"line_number":58,"context_line":"request, nor do we plan to control what goes on inside containers."},{"line_number":59,"context_line":"They get the same metadata provided to them as virtual machines, to do"},{"line_number":60,"context_line":"with as they see fit."},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Bare Metal Servers"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_a96a697f","line":59,"in_reply_to":"1aca2d91_0b4cfa6e","updated":"2015-05-08 19:52:51.000000000","message":"+1, that has always been an important distinction for me: Nova stops at the boundary to the guest. That helps delineate any container activities we may dip our toe into. Whether it needs to be fleshed out too much more or not, I dunno. However, I think it\u0027s a good thing to mention.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":56,"context_line":""},{"line_number":57,"context_line":"For example, Nova is not designed to spin up new containers for every apache"},{"line_number":58,"context_line":"request, nor do we plan to control what goes on inside containers."},{"line_number":59,"context_line":"They get the same metadata provided to them as virtual machines, to do"},{"line_number":60,"context_line":"with as they see fit."},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"Bare Metal Servers"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_0b4cfa6e","line":59,"in_reply_to":"1aca2d91_60a6ffc2","updated":"2015-05-07 07:28:53.000000000","message":"Yes, true.  But I want to get *something* in sooner.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":5754,"name":"Alex Xu","email":"hejie.xu@intel.com","username":"xuhj"},"change_message_id":"55e0dcc5c12a0399bd88ee0880e0dfa51c9aab47","unresolved":false,"context_lines":[{"line_number":70,"context_line":""},{"line_number":71,"context_line":"While currently there are operations that are a fundamental part of our"},{"line_number":72,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":73,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":74,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":75,"context_line":"it is expected that gap with shrink over time."},{"line_number":76,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_216b5720","line":73,"updated":"2015-05-07 08:59:54.000000000","message":"s/does don\u0027t/don\u0027t/?","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":70,"context_line":""},{"line_number":71,"context_line":"While currently there are operations that are a fundamental part of our"},{"line_number":72,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":73,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":74,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":75,"context_line":"it is expected that gap with shrink over time."},{"line_number":76,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_c9e6c514","line":73,"in_reply_to":"1aca2d91_216b5720","updated":"2015-05-08 19:52:51.000000000","message":"Actually, it should be \"does not\"","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":5754,"name":"Alex Xu","email":"hejie.xu@intel.com","username":"xuhj"},"change_message_id":"55e0dcc5c12a0399bd88ee0880e0dfa51c9aab47","unresolved":false,"context_lines":[{"line_number":71,"context_line":"While currently there are operations that are a fundamental part of our"},{"line_number":72,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":73,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":74,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":75,"context_line":"it is expected that gap with shrink over time."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Driver Parity"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_815a2b82","line":74,"updated":"2015-05-07 08:59:54.000000000","message":"double \u0027our\u0027 at here.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":72,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":73,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":74,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":75,"context_line":"it is expected that gap with shrink over time."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Driver Parity"},{"line_number":78,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_0c5babf7","line":75,"updated":"2015-05-08 19:52:51.000000000","message":"I think the ironic driver is a stretch and somewhat of a special case. Maybe that\u0027s what you\u0027re trying to say here. I just hope it doesn\u0027t get interpreted as \"other weirdo drivers are okay as long as they\u0027re weird enough to not have $features be relevant\"","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":75,"context_line":"it is expected that gap with shrink over time."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Driver Parity"},{"line_number":78,"context_line":"--------------"},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"When I say driver, I am talking about the code that is used to control the"},{"line_number":81,"context_line":"specific type of compute resource a specific nova-compute node is controlling,"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_40b0a3dd","line":78,"updated":"2015-05-06 19:10:06.000000000","message":"This section seems like its really long and doesn\u0027t say much","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":75,"context_line":"it is expected that gap with shrink over time."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Driver Parity"},{"line_number":78,"context_line":"--------------"},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"When I say driver, I am talking about the code that is used to control the"},{"line_number":81,"context_line":"specific type of compute resource a specific nova-compute node is controlling,"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_4bb0b243","line":78,"in_reply_to":"1aca2d91_40b0a3dd","updated":"2015-05-07 07:28:53.000000000","message":"It can be reworded for sure. Its important though.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":77,"context_line":"Driver Parity"},{"line_number":78,"context_line":"--------------"},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"When I say driver, I am talking about the code that is used to control the"},{"line_number":81,"context_line":"specific type of compute resource a specific nova-compute node is controlling,"},{"line_number":82,"context_line":"such as libvirt, ironic or a hypervisor specific driver."},{"line_number":83,"context_line":"Its possible for a single nova deployment to use multiple different drivers."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_a073f74c","line":80,"updated":"2015-05-06 19:10:06.000000000","message":"First person voice?","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":77,"context_line":"Driver Parity"},{"line_number":78,"context_line":"--------------"},{"line_number":79,"context_line":""},{"line_number":80,"context_line":"When I say driver, I am talking about the code that is used to control the"},{"line_number":81,"context_line":"specific type of compute resource a specific nova-compute node is controlling,"},{"line_number":82,"context_line":"such as libvirt, ironic or a hypervisor specific driver."},{"line_number":83,"context_line":"Its possible for a single nova deployment to use multiple different drivers."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_eb5f7e0c","line":80,"in_reply_to":"1aca2d91_a073f74c","updated":"2015-05-07 07:28:53.000000000","message":"sure, just trying to get a starting point in.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":90,"context_line":""},{"line_number":91,"context_line":"Following on from that, should a feature only be added to one of the drivers,"},{"line_number":92,"context_line":"we must make every effort to ensure another driver could be implemented to"},{"line_number":93,"context_line":"match that behavior."},{"line_number":94,"context_line":""},{"line_number":95,"context_line":"The reason for this approach, is that the Nova API should look and feel the"},{"line_number":96,"context_line":"same no matter what OpenStack deployment you are using. Its possible you have"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_ccf7f389","line":93,"updated":"2015-05-08 19:52:51.000000000","message":"I wonder if we should say something about importing existing resources? We\u0027ve had at least one proposed driver that didn\u0027t implement spawn() or delete(), and instead just \"imported\" VMs already running on the hypervisor into the nova database on startup.\n\nNova manages only resources it creates, and that\u0027s important to call out. Not sure if it belongs here, or if it\u0027s too specific, but I thought I\u0027d mention it.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e31879f5720e6f8f3b4a2dd4267aa424a131b4ab","unresolved":false,"context_lines":[{"line_number":90,"context_line":""},{"line_number":91,"context_line":"Following on from that, should a feature only be added to one of the drivers,"},{"line_number":92,"context_line":"we must make every effort to ensure another driver could be implemented to"},{"line_number":93,"context_line":"match that behavior."},{"line_number":94,"context_line":""},{"line_number":95,"context_line":"The reason for this approach, is that the Nova API should look and feel the"},{"line_number":96,"context_line":"same no matter what OpenStack deployment you are using. Its possible you have"}],"source_content_type":"text/x-rst","patch_set":8,"id":"9af37de9_4463421b","line":93,"in_reply_to":"1aca2d91_9d8b6368","updated":"2015-05-11 09:14:43.000000000","message":"+1","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"07742075dc7eaccd0780940d12a0c2eaa9cde078","unresolved":false,"context_lines":[{"line_number":90,"context_line":""},{"line_number":91,"context_line":"Following on from that, should a feature only be added to one of the drivers,"},{"line_number":92,"context_line":"we must make every effort to ensure another driver could be implemented to"},{"line_number":93,"context_line":"match that behavior."},{"line_number":94,"context_line":""},{"line_number":95,"context_line":"The reason for this approach, is that the Nova API should look and feel the"},{"line_number":96,"context_line":"same no matter what OpenStack deployment you are using. Its possible you have"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_9d8b6368","line":93,"in_reply_to":"1aca2d91_ccf7f389","updated":"2015-05-08 20:59:57.000000000","message":"I think it\u0027s important to call out that Nova manages only resources it creates.  People have certainly had the impression, or desire, that it\u0027s not true.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":103,"context_line":"on demand compute resources, rather than expose all features of all"},{"line_number":104,"context_line":"hypervisors. Where the details of the underlying hypervisor leak through"},{"line_number":105,"context_line":"our APIs, we have failed in this goal, and we should work towards better"},{"line_number":106,"context_line":"abstractions that are more interoperable."},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"Deployment and Packaging"},{"line_number":109,"context_line":"-------------------------"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_0c85eb05","line":106,"updated":"2015-05-08 19:52:51.000000000","message":"So, in the past I\u0027ve been a big proponent of a consistent baseline of features for virt drivers. Partly because it ensures the above statement of \"feel\" from the view of an API user, but also because it helps to provide consistence of what an openstack guest looks like. The in-guest environment (metadata server, configdrive, etc) is important. Finally, the bar helps us define whether something belongs in nova or not. If we can only implement half of the supposed bottom line functionality, it\u0027s probably not really a virt driver, at least in terms of what nova considers.\n\nI know some people want no bar at all, but I wonder if we can call out at least the fact that there should be a bar that all drivers should meet here?","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e31879f5720e6f8f3b4a2dd4267aa424a131b4ab","unresolved":false,"context_lines":[{"line_number":103,"context_line":"on demand compute resources, rather than expose all features of all"},{"line_number":104,"context_line":"hypervisors. Where the details of the underlying hypervisor leak through"},{"line_number":105,"context_line":"our APIs, we have failed in this goal, and we should work towards better"},{"line_number":106,"context_line":"abstractions that are more interoperable."},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"Deployment and Packaging"},{"line_number":109,"context_line":"-------------------------"}],"source_content_type":"text/x-rst","patch_set":8,"id":"9af37de9_24683ef5","line":106,"in_reply_to":"1aca2d91_0c85eb05","updated":"2015-05-11 09:14:43.000000000","message":"+1","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":124,"context_line":"Our mission includes the text \"massively scalable\". Lets discuss what that"},{"line_number":125,"context_line":"means."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":128,"context_line":"servers (which implies a number of active compute nodes)."},{"line_number":129,"context_line":"Its possible to have a small cloud with lots of requests for very short"},{"line_number":130,"context_line":"lived VMs, or a large cloud with lots of longer lived VMs."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_c0db931a","line":127,"updated":"2015-05-06 19:10:06.000000000","message":"I still count 3:\n * API requests\n * Physical boxes\n * Servers/Instances","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"e31879f5720e6f8f3b4a2dd4267aa424a131b4ab","unresolved":false,"context_lines":[{"line_number":124,"context_line":"Our mission includes the text \"massively scalable\". Lets discuss what that"},{"line_number":125,"context_line":"means."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":128,"context_line":"servers (which implies a number of active compute nodes)."},{"line_number":129,"context_line":"Its possible to have a small cloud with lots of requests for very short"},{"line_number":130,"context_line":"lived VMs, or a large cloud with lots of longer lived VMs."}],"source_content_type":"text/x-rst","patch_set":8,"id":"9af37de9_a4742ecf","line":127,"in_reply_to":"1aca2d91_2c97873b","updated":"2015-05-11 09:14:43.000000000","message":"very true, sadly.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"85469dfb4767f23ee44ff1213aba12492d673581","unresolved":false,"context_lines":[{"line_number":124,"context_line":"Our mission includes the text \"massively scalable\". Lets discuss what that"},{"line_number":125,"context_line":"means."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":128,"context_line":"servers (which implies a number of active compute nodes)."},{"line_number":129,"context_line":"Its possible to have a small cloud with lots of requests for very short"},{"line_number":130,"context_line":"lived VMs, or a large cloud with lots of longer lived VMs."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_e6b1b048","line":127,"in_reply_to":"1aca2d91_ab874660","updated":"2015-05-08 18:31:49.000000000","message":"Sure they scale linearly and together. but they can be an order of magnitude or two different in scale. and one can be a limiting factor while the other isn\u0027t. etc.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":124,"context_line":"Our mission includes the text \"massively scalable\". Lets discuss what that"},{"line_number":125,"context_line":"means."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":128,"context_line":"servers (which implies a number of active compute nodes)."},{"line_number":129,"context_line":"Its possible to have a small cloud with lots of requests for very short"},{"line_number":130,"context_line":"lived VMs, or a large cloud with lots of longer lived VMs."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_ab874660","line":127,"in_reply_to":"1aca2d91_c0db931a","updated":"2015-05-07 07:28:53.000000000","message":"Physical boxes and instances are so highly correlated, so I don\u0027t like making them separate. The scale linearly and together.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":124,"context_line":"Our mission includes the text \"massively scalable\". Lets discuss what that"},{"line_number":125,"context_line":"means."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Nova has two levels of scale. Numbers of API requests and numbers of active"},{"line_number":128,"context_line":"servers (which implies a number of active compute nodes)."},{"line_number":129,"context_line":"Its possible to have a small cloud with lots of requests for very short"},{"line_number":130,"context_line":"lived VMs, or a large cloud with lots of longer lived VMs."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_2c97873b","line":127,"in_reply_to":"1aca2d91_e6b1b048","updated":"2015-05-08 19:52:51.000000000","message":"Agreed, they don\u0027t scale together in the same way with the vmware driver, nor with containers, as they do with something like libvirt or xen.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":152,"context_line":"---------------------"},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"Our upgrade plan is to concentrate on upgrades from N-1 to the Nth release."},{"line_number":155,"context_line":"So for someone running juno, they would have to upgrade kilo before upgrading"},{"line_number":156,"context_line":"to liberty."},{"line_number":157,"context_line":"This is designed to balance the need for a smooth upgrade, against having to"},{"line_number":158,"context_line":"keep maintaining the compatibility code to make that upgrade possible."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_cc3ed327","line":155,"range":{"start_line":155,"start_character":48,"end_line":155,"end_character":55},"updated":"2015-05-08 19:52:51.000000000","message":"should be \"upgrade to\"","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":159,"context_line":"We talk about this approach as users consuming the stable branch."},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"In addition, we also support users upgrading from the master branch."},{"line_number":162,"context_line":"Technically, between any two between any two commits within the"},{"line_number":163,"context_line":"same release cycle. In certain cases, when cross release boundaries, you must"},{"line_number":164,"context_line":"upgrade to the stable branch, before then upgrading to the tip of master."},{"line_number":165,"context_line":"This is to support those that are doing some level of"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_60303f57","line":162,"updated":"2015-05-06 19:10:06.000000000","message":"Is this too much detail?","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":159,"context_line":"We talk about this approach as users consuming the stable branch."},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"In addition, we also support users upgrading from the master branch."},{"line_number":162,"context_line":"Technically, between any two between any two commits within the"},{"line_number":163,"context_line":"same release cycle. In certain cases, when cross release boundaries, you must"},{"line_number":164,"context_line":"upgrade to the stable branch, before then upgrading to the tip of master."},{"line_number":165,"context_line":"This is to support those that are doing some level of"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_cc25b309","line":162,"in_reply_to":"1aca2d91_2b945689","updated":"2015-05-08 19:52:51.000000000","message":"I think it\u0027s good to call out.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":159,"context_line":"We talk about this approach as users consuming the stable branch."},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"In addition, we also support users upgrading from the master branch."},{"line_number":162,"context_line":"Technically, between any two between any two commits within the"},{"line_number":163,"context_line":"same release cycle. In certain cases, when cross release boundaries, you must"},{"line_number":164,"context_line":"upgrade to the stable branch, before then upgrading to the tip of master."},{"line_number":165,"context_line":"This is to support those that are doing some level of"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_2b945689","line":162,"in_reply_to":"1aca2d91_60303f57","updated":"2015-05-07 07:28:53.000000000","message":"It can become important, but maybe.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":160,"context_line":""},{"line_number":161,"context_line":"In addition, we also support users upgrading from the master branch."},{"line_number":162,"context_line":"Technically, between any two between any two commits within the"},{"line_number":163,"context_line":"same release cycle. In certain cases, when cross release boundaries, you must"},{"line_number":164,"context_line":"upgrade to the stable branch, before then upgrading to the tip of master."},{"line_number":165,"context_line":"This is to support those that are doing some level of"},{"line_number":166,"context_line":"\"Continuous Deployment\" from the tip of master into production."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_4c8083c0","line":163,"range":{"start_line":163,"start_character":43,"end_line":163,"end_character":48},"updated":"2015-05-08 19:52:51.000000000","message":"\"crossing\" ?","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":227,"context_line":"----------------------"},{"line_number":228,"context_line":""},{"line_number":229,"context_line":"Nova is a low level infrastructure API. It is plumbing upon which richer"},{"line_number":230,"context_line":"ideas can be built. Heat and Magnum being a great examples of that."},{"line_number":231,"context_line":""},{"line_number":232,"context_line":"While we have some APIs that could be considered orchestration, and we must"},{"line_number":233,"context_line":"continue to maintain those, we do not intend to add any more APIs that do"}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_ac4cb7c8","line":230,"range":{"start_line":230,"start_character":42,"end_line":230,"end_character":43},"updated":"2015-05-08 19:52:51.000000000","message":"s/a//","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":242,"context_line":""},{"line_number":243,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"},{"line_number":244,"context_line":"for compute resources compared with the amount of compute resource thats"},{"line_number":245,"context_line":"currently available. We generally assume a level of spare capacity is"},{"line_number":246,"context_line":"maintained for future requests. As such our quota system is more"},{"line_number":247,"context_line":"focused on limiting the current level of resource usage, rather than"},{"line_number":248,"context_line":"ensuring a fair balance of resources between all incoming requests."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_405ae384","line":245,"updated":"2015-05-06 19:10:06.000000000","message":"\"Spare Capacity\" is a literally a waste of money. This isn\u0027t something we should assume.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":242,"context_line":""},{"line_number":243,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"},{"line_number":244,"context_line":"for compute resources compared with the amount of compute resource thats"},{"line_number":245,"context_line":"currently available. We generally assume a level of spare capacity is"},{"line_number":246,"context_line":"maintained for future requests. As such our quota system is more"},{"line_number":247,"context_line":"focused on limiting the current level of resource usage, rather than"},{"line_number":248,"context_line":"ensuring a fair balance of resources between all incoming requests."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_6c657f1c","line":245,"in_reply_to":"1aca2d91_067f0c67","updated":"2015-05-08 19:52:51.000000000","message":"It doesn\u0027t close the door, it just closes the door on selling 100% of spare capacity.\n\nPersonally, I don\u0027t think that IaaS makes much sense if you run at 100.0% utilization all the time. If nothing is free, nothing can move, and thus you\u0027re pretty static. You might spike to 100% at times, but that\u0027s not really what this point is forbidding. IMHO, it\u0027s making \"running 100% packed hypervisors with all CPUs pinned and zero free bytes of memory\" clearly part of VMware\u0027s business model and not something nova is going to optimize for or go to great lengths to support.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":5441,"name":"Andrew Laski","email":"andrew@lascii.com","username":"alaski"},"change_message_id":"c0e06ff1b5d006300989bd5a9a5a703354c0d3e0","unresolved":false,"context_lines":[{"line_number":242,"context_line":""},{"line_number":243,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"},{"line_number":244,"context_line":"for compute resources compared with the amount of compute resource thats"},{"line_number":245,"context_line":"currently available. We generally assume a level of spare capacity is"},{"line_number":246,"context_line":"maintained for future requests. As such our quota system is more"},{"line_number":247,"context_line":"focused on limiting the current level of resource usage, rather than"},{"line_number":248,"context_line":"ensuring a fair balance of resources between all incoming requests."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_4ad1dacc","line":245,"in_reply_to":"1aca2d91_405ae384","updated":"2015-05-06 21:53:51.000000000","message":"I think parts of Nova were built with an assumption that there were plenty of resources available, the scheduler being the main one.  It relies on there being enough capacity that races are relatively rare.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":242,"context_line":""},{"line_number":243,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"},{"line_number":244,"context_line":"for compute resources compared with the amount of compute resource thats"},{"line_number":245,"context_line":"currently available. We generally assume a level of spare capacity is"},{"line_number":246,"context_line":"maintained for future requests. As such our quota system is more"},{"line_number":247,"context_line":"focused on limiting the current level of resource usage, rather than"},{"line_number":248,"context_line":"ensuring a fair balance of resources between all incoming requests."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_abd50644","line":245,"in_reply_to":"1aca2d91_4ad1dacc","updated":"2015-05-07 07:28:53.000000000","message":"For a \"dynamic\" cloud you need \"spare capacity\" as people scale up and down (well out and in). Its a nasty secret that the more dynamic folks are the more \"spare capacity\" you need.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"85469dfb4767f23ee44ff1213aba12492d673581","unresolved":false,"context_lines":[{"line_number":242,"context_line":""},{"line_number":243,"context_line":"Nova is not currently optimized for dealing with a larger number of requests"},{"line_number":244,"context_line":"for compute resources compared with the amount of compute resource thats"},{"line_number":245,"context_line":"currently available. We generally assume a level of spare capacity is"},{"line_number":246,"context_line":"maintained for future requests. As such our quota system is more"},{"line_number":247,"context_line":"focused on limiting the current level of resource usage, rather than"},{"line_number":248,"context_line":"ensuring a fair balance of resources between all incoming requests."}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_067f0c67","line":245,"in_reply_to":"1aca2d91_abd50644","updated":"2015-05-08 18:31:49.000000000","message":"Agreed, but saying this closes the door for spot instances.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"58e38fcd06a469019b6fade8b586a741489919f8","unresolved":false,"context_lines":[{"line_number":250,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":251,"context_line":"batch job processing."},{"line_number":252,"context_line":"But the current focus is on how to layer such an abstraction on top of the"},{"line_number":253,"context_line":"basic primitives Nova currently provides, possibly adding additional APIs"},{"line_number":254,"context_line":"where that makes good sense. Should this turn out to be impractical, we may"},{"line_number":255,"context_line":"have to revise our approach."},{"line_number":256,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_e06a0f1a","line":253,"updated":"2015-05-06 19:10:06.000000000","message":"So I don\u0027t think this is really viable. I would love to see OpenStack support the equivalent of amazon spot instances. But I don\u0027t see how that can be done \u0027on top\u0027 of Nova.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2a7fe7d904b7c43283abd3c175afef7940b8af19","unresolved":false,"context_lines":[{"line_number":250,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":251,"context_line":"batch job processing."},{"line_number":252,"context_line":"But the current focus is on how to layer such an abstraction on top of the"},{"line_number":253,"context_line":"basic primitives Nova currently provides, possibly adding additional APIs"},{"line_number":254,"context_line":"where that makes good sense. Should this turn out to be impractical, we may"},{"line_number":255,"context_line":"have to revise our approach."},{"line_number":256,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_4c1f636c","line":253,"in_reply_to":"1aca2d91_66eb6014","updated":"2015-05-08 19:52:51.000000000","message":"I think what is here is pretty accurate. There is a spec up right now that aims to provide a queuing mechanism in Nova, where you can submit a build request that will sit in a queue until something is removed to make room for you. That\u0027s not spot instances, that\u0027s batch processing of 100.0% system capacity. When you do that in nova, it means we need to get into fairness and priority queuing, and then the next step is killing things that are running when something higher priority comes in.\n\nAll of that can be done on top of nova, and should be, IMHO.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"85469dfb4767f23ee44ff1213aba12492d673581","unresolved":false,"context_lines":[{"line_number":250,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":251,"context_line":"batch job processing."},{"line_number":252,"context_line":"But the current focus is on how to layer such an abstraction on top of the"},{"line_number":253,"context_line":"basic primitives Nova currently provides, possibly adding additional APIs"},{"line_number":254,"context_line":"where that makes good sense. Should this turn out to be impractical, we may"},{"line_number":255,"context_line":"have to revise our approach."},{"line_number":256,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_66eb6014","line":253,"in_reply_to":"1aca2d91_6b5daee8","updated":"2015-05-08 18:31:49.000000000","message":"works for me","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"958946a1988c1820c59f529109468ced9d2e5855","unresolved":false,"context_lines":[{"line_number":250,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":251,"context_line":"batch job processing."},{"line_number":252,"context_line":"But the current focus is on how to layer such an abstraction on top of the"},{"line_number":253,"context_line":"basic primitives Nova currently provides, possibly adding additional APIs"},{"line_number":254,"context_line":"where that makes good sense. Should this turn out to be impractical, we may"},{"line_number":255,"context_line":"have to revise our approach."},{"line_number":256,"context_line":""}],"source_content_type":"text/x-rst","patch_set":8,"id":"1aca2d91_6b5daee8","line":253,"in_reply_to":"1aca2d91_e06a0f1a","updated":"2015-05-07 07:28:53.000000000","message":"I don\u0027t mean spot instances here, that should be part of Nova, and is about monetising \"spare\" capacity.\n\nI mean balanced quotas across a large list of \"jobs\". Maybe I should call out spot instances here.","commit_id":"ec1dc9b11d76c561615c50a0b5bbdf9fe5f83c6f"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2085297e132a1dae93c5584411b113713f4542df","unresolved":false,"context_lines":[{"line_number":46,"context_line":"Virtual Servers"},{"line_number":47,"context_line":"****************"},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"Nova was originally focused purely of providing access to virtual servers"},{"line_number":50,"context_line":"running on a variety of different hypervisors."},{"line_number":51,"context_line":"The majority of users use Nova only to provide access to virtual servers"},{"line_number":52,"context_line":"from a single hypervisor, however, its possibly to have a Nova deployment"}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_7701c804","line":49,"range":{"start_line":49,"start_character":35,"end_line":49,"end_character":37},"updated":"2015-05-08 19:57:27.000000000","message":"s/of/on/","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2085297e132a1dae93c5584411b113713f4542df","unresolved":false,"context_lines":[{"line_number":49,"context_line":"Nova was originally focused purely of providing access to virtual servers"},{"line_number":50,"context_line":"running on a variety of different hypervisors."},{"line_number":51,"context_line":"The majority of users use Nova only to provide access to virtual servers"},{"line_number":52,"context_line":"from a single hypervisor, however, its possibly to have a Nova deployment"},{"line_number":53,"context_line":"includes multiple different types of hypervisors, while at the same time"},{"line_number":54,"context_line":"offering containers and bare metal servers."},{"line_number":55,"context_line":""}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_170cd439","line":52,"range":{"start_line":52,"start_character":39,"end_line":52,"end_character":47},"updated":"2015-05-08 19:57:27.000000000","message":"\"possible\"","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2085297e132a1dae93c5584411b113713f4542df","unresolved":false,"context_lines":[{"line_number":50,"context_line":"running on a variety of different hypervisors."},{"line_number":51,"context_line":"The majority of users use Nova only to provide access to virtual servers"},{"line_number":52,"context_line":"from a single hypervisor, however, its possibly to have a Nova deployment"},{"line_number":53,"context_line":"includes multiple different types of hypervisors, while at the same time"},{"line_number":54,"context_line":"offering containers and bare metal servers."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"Containers"}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_9718e476","line":53,"range":{"start_line":53,"start_character":0,"end_line":53,"end_character":8},"updated":"2015-05-08 19:57:27.000000000","message":"\"include\"","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"ec81c931a9e9dd3bb0f44c2ed6f9311d64be20df","unresolved":false,"context_lines":[{"line_number":51,"context_line":"The majority of users use Nova only to provide access to virtual servers"},{"line_number":52,"context_line":"from a single hypervisor, however, its possibly to have a Nova deployment"},{"line_number":53,"context_line":"includes multiple different types of hypervisors, while at the same time"},{"line_number":54,"context_line":"offering containers and bare metal servers."},{"line_number":55,"context_line":""},{"line_number":56,"context_line":"Containers"},{"line_number":57,"context_line":"***********"}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_a66a987f","line":54,"updated":"2015-05-08 18:39:25.000000000","message":"Nicely put!","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"ec81c931a9e9dd3bb0f44c2ed6f9311d64be20df","unresolved":false,"context_lines":[{"line_number":60,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":61,"context_line":"on top of Nova."},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"The Magnum project has the following mission:"},{"line_number":64,"context_line":"To provide a set of services for management of application containers"},{"line_number":65,"context_line":"in a multitenant cloud environment."},{"line_number":66,"context_line":""}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_e6fbf0bf","line":63,"updated":"2015-05-08 18:39:25.000000000","message":"I still stand by we don\u0027t need to call out magnum here. One can use straight kubranetes, swarm etc. or other things if they want as well","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"62fce8ab335db5d7b37e0c17629ed62544096fe3","unresolved":false,"context_lines":[{"line_number":60,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":61,"context_line":"on top of Nova."},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"The Magnum project has the following mission:"},{"line_number":64,"context_line":"To provide a set of services for management of application containers"},{"line_number":65,"context_line":"in a multitenant cloud environment."},{"line_number":66,"context_line":""}],"source_content_type":"text/x-rst","patch_set":12,"id":"9af37de9_a479ce31","line":63,"in_reply_to":"1aca2d91_d739bcd8","updated":"2015-05-11 09:15:32.000000000","message":"OK, cool. Still adjusting to the big tent I guess.","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2085297e132a1dae93c5584411b113713f4542df","unresolved":false,"context_lines":[{"line_number":60,"context_line":"The Magnum project intends to deliver a good container experience built"},{"line_number":61,"context_line":"on top of Nova."},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"The Magnum project has the following mission:"},{"line_number":64,"context_line":"To provide a set of services for management of application containers"},{"line_number":65,"context_line":"in a multitenant cloud environment."},{"line_number":66,"context_line":""}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_d739bcd8","line":63,"in_reply_to":"1aca2d91_e6fbf0bf","updated":"2015-05-08 19:57:27.000000000","message":"I agree, we should drop magnum from being mentioned. See my previous \"big tent\" justification :)","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"2085297e132a1dae93c5584411b113713f4542df","unresolved":false,"context_lines":[{"line_number":144,"context_line":"been deployed into production, and as such cannot be reverted."},{"line_number":145,"context_line":"In a similar way, a database migration may have been deployed."},{"line_number":146,"context_line":""},{"line_number":147,"context_line":"Any commit that will affect and upgrade gets the UpgradeImpact tag added to"},{"line_number":148,"context_line":"the commit message, so there is no requirement to wait for release notes."},{"line_number":149,"context_line":""},{"line_number":150,"context_line":"Don\u0027t break exisiting users"}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_b76080d0","line":147,"range":{"start_line":147,"start_character":28,"end_line":147,"end_character":31},"updated":"2015-05-08 19:57:27.000000000","message":"s/and/an/","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":1849,"name":"Joe Gordon","email":"joe.gordon0@gmail.com","username":"jogo"},"change_message_id":"ec81c931a9e9dd3bb0f44c2ed6f9311d64be20df","unresolved":false,"context_lines":[{"line_number":277,"context_line":"resource usage, rather than ensuring a fair balance of resources between all"},{"line_number":278,"context_line":"incoming requests."},{"line_number":279,"context_line":"This doesn\u0027t exclude adding features to make better use of spare capacity,"},{"line_number":280,"context_line":"such a \"spot instances\"."},{"line_number":281,"context_line":""},{"line_number":282,"context_line":"There have been discussions around how to change Nova to work better for"},{"line_number":283,"context_line":"batch job processing."}],"source_content_type":"text/x-rst","patch_set":12,"id":"1aca2d91_66a360b3","line":280,"updated":"2015-05-08 18:39:25.000000000","message":"thanks","commit_id":"d6105a7b6eeb2ce3715d7e1384906022e427a55a"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"8800be9efaec88b75306994890d820045bd1fa71","unresolved":false,"context_lines":[{"line_number":148,"context_line":"Any commit that will affect an upgrade gets the UpgradeImpact tag added to"},{"line_number":149,"context_line":"the commit message, so there is no requirement to wait for release notes."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"Don\u0027t break exisiting users"},{"line_number":152,"context_line":"****************************"},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"As a community we are aiming towards a smooth upgrade process, where users"}],"source_content_type":"text/x-rst","patch_set":14,"id":"9a0c5dc1_980f33e1","line":151,"range":{"start_line":151,"start_character":12,"end_line":151,"end_character":21},"updated":"2015-05-12 22:22:47.000000000","message":"\"existing\" is misspelled","commit_id":"415ed85c6a6a4089566dca05718b42e1651d92d2"}],"doc/source/devref/what_we_dont_do.rst":[{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"98b7f39aa43f54ecb12bbbe408272f21cd2e8bb5","unresolved":false,"context_lines":[{"line_number":11,"context_line":"      License for the specific language governing permissions and limitations"},{"line_number":12,"context_line":"      under the License."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"What Nova does\u0027t do"},{"line_number":15,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"There have been many discussion around what is not in scope."}],"source_content_type":"text/x-rst","patch_set":1,"id":"5ad0253b_b8f1eccf","line":14,"updated":"2015-04-29 13:41:12.000000000","message":"s/does\u0027t/doesn\u0027t","commit_id":"d1e61487011e11fa70f57e7ce9ec87e43c8c7ab9"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"98b7f39aa43f54ecb12bbbe408272f21cd2e8bb5","unresolved":false,"context_lines":[{"line_number":14,"context_line":"What Nova does\u0027t do"},{"line_number":15,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"There have been many discussion around what is not in scope."},{"line_number":18,"context_line":""},{"line_number":19,"context_line":"This is a living document to help record where we agree about what Nova"},{"line_number":20,"context_line":"should not be doing, and why."}],"source_content_type":"text/x-rst","patch_set":1,"id":"5ad0253b_98ee30eb","line":17,"updated":"2015-04-29 13:41:12.000000000","message":"s/discussion/discussions","commit_id":"d1e61487011e11fa70f57e7ce9ec87e43c8c7ab9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"b3b97937a74919273561838ba36ac6586f1b1693","unresolved":false,"context_lines":[{"line_number":27,"context_line":"To implement services and associated libraries to provide massively"},{"line_number":28,"context_line":"scalable, on demand, self service access to compute resources, including"},{"line_number":29,"context_line":"bare metal, virtual machines, and containers."},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"Containers"},{"line_number":32,"context_line":"----------"},{"line_number":33,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"5ad0253b_98e4f03b","line":30,"updated":"2015-04-29 13:36:53.000000000","message":"Should we perhaps provide the mission statement ? I know it can be given elsewhere, but it would be cool if we would the statement to explain why we do/don\u0027t","commit_id":"d1e61487011e11fa70f57e7ce9ec87e43c8c7ab9"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"b2db46063be06050c894e6408b5767dbca663152","unresolved":false,"context_lines":[{"line_number":27,"context_line":"To implement services and associated libraries to provide massively"},{"line_number":28,"context_line":"scalable, on demand, self service access to compute resources, including"},{"line_number":29,"context_line":"bare metal, virtual machines, and containers."},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"Containers"},{"line_number":32,"context_line":"----------"},{"line_number":33,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3acd31a7_d996cc4b","line":30,"in_reply_to":"5ad0253b_98e4f03b","updated":"2015-05-01 12:06:18.000000000","message":"That is the mission statement.","commit_id":"d1e61487011e11fa70f57e7ce9ec87e43c8c7ab9"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"98b7f39aa43f54ecb12bbbe408272f21cd2e8bb5","unresolved":false,"context_lines":[{"line_number":35,"context_line":"on demand virtual machines."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"However we are not designed to spin up new containers for every apache"},{"line_number":38,"context_line":"request, nor do we plan to control what goes on inside a containers."},{"line_number":39,"context_line":"They get the same metadata provided to them as virtual machines, to do"},{"line_number":40,"context_line":"with as they see fit."},{"line_number":41,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"5ad0253b_78ba34e3","line":38,"updated":"2015-04-29 13:41:12.000000000","message":"s/inside a containers/inside a container","commit_id":"d1e61487011e11fa70f57e7ce9ec87e43c8c7ab9"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"b3b97937a74919273561838ba36ac6586f1b1693","unresolved":false,"context_lines":[{"line_number":52,"context_line":"-------------"},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"TODO..."},{"line_number":55,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"5ad0253b_78d0f453","line":55,"updated":"2015-04-29 13:36:53.000000000","message":"Resource Tracking :-)","commit_id":"d1e61487011e11fa70f57e7ce9ec87e43c8c7ab9"},{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"03b9111f0671f61a5f916ddc1728ac573279d41d","unresolved":false,"context_lines":[{"line_number":38,"context_line":"Containers"},{"line_number":39,"context_line":"----------"},{"line_number":40,"context_line":""},{"line_number":41,"context_line":"Nova allows you to use containers in a similar way to how you would use"},{"line_number":42,"context_line":"on demand virtual machines. We want to maintain this distinction, so we"},{"line_number":43,"context_line":"maintain the integrity and usefulness of the existing Nova API."},{"line_number":44,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_a6040fc0","line":41,"updated":"2015-05-01 10:03:14.000000000","message":"when yoiu say containers are you referring to lxc or docker?\nit is a huge problem that the docker drivers are not in tree. up side on that is that they can actually do development and not be blocked by community reviews...","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"b99186de57739a0aa8d7b64f9f5b12ef8e689f7c","unresolved":false,"context_lines":[{"line_number":38,"context_line":"Containers"},{"line_number":39,"context_line":"----------"},{"line_number":40,"context_line":""},{"line_number":41,"context_line":"Nova allows you to use containers in a similar way to how you would use"},{"line_number":42,"context_line":"on demand virtual machines. We want to maintain this distinction, so we"},{"line_number":43,"context_line":"maintain the integrity and usefulness of the existing Nova API."},{"line_number":44,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_9977343a","line":41,"in_reply_to":"3acd31a7_a6040fc0","updated":"2015-05-01 12:08:44.000000000","message":"I am specifically not defining the technology here.\n\nI am happy for the docker driver to be in tree, if they meet these guidelines, and if they can get their CI in place as requested, and get the appropriate spec approved for liberty. But thats a separate topic.","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":5511,"name":"Nikola Dipanov","email":"ndipanov@redhat.com","username":"ndipanov"},"change_message_id":"49fa68011dde76b813041438626897132e6420d4","unresolved":false,"context_lines":[{"line_number":64,"context_line":"virtual machine abstraction that are not currently available in ironic,"},{"line_number":65,"context_line":"such as attaching iSCSI volumes, it does don\u0027t fundamentally change the"},{"line_number":66,"context_line":"semantics our our API, and as such is a suitable Nova driver. Moreover,"},{"line_number":67,"context_line":"it is expected that gap with shrink over time."},{"line_number":68,"context_line":""},{"line_number":69,"context_line":"Driver Parity"},{"line_number":70,"context_line":"--------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_5c4bca87","line":67,"updated":"2015-05-01 14:32:12.000000000","message":"Not sure there is a need for the last sentence at all in the context of this document, but that\u0027s only a nit really.","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"03b9111f0671f61a5f916ddc1728ac573279d41d","unresolved":false,"context_lines":[{"line_number":68,"context_line":""},{"line_number":69,"context_line":"Driver Parity"},{"line_number":70,"context_line":"--------------"},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"The key tenant of driver parity is that if a feature is supported in a driver,"},{"line_number":73,"context_line":"it should feel the same to users, as if they where using any of the other"},{"line_number":74,"context_line":"drivers that also support that feature. The exception is that, if possible for"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_e621e74e","line":71,"updated":"2015-05-01 10:03:14.000000000","message":"It would be nice for the drivers in tree to be able to get reviews for parity features - examples are features for libvirt being approved in I and the same features waiting for review for other drivers in J, K, and now waiting in L. The code in most cases was ready in J.\nSo essentially we do not have driver parity. I understand the mission statement but in the trenches things are not like this. If we can fix it great!","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"b99186de57739a0aa8d7b64f9f5b12ef8e689f7c","unresolved":false,"context_lines":[{"line_number":68,"context_line":""},{"line_number":69,"context_line":"Driver Parity"},{"line_number":70,"context_line":"--------------"},{"line_number":71,"context_line":""},{"line_number":72,"context_line":"The key tenant of driver parity is that if a feature is supported in a driver,"},{"line_number":73,"context_line":"it should feel the same to users, as if they where using any of the other"},{"line_number":74,"context_line":"drivers that also support that feature. The exception is that, if possible for"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_7972184b","line":71,"in_reply_to":"3acd31a7_e621e74e","updated":"2015-05-01 12:08:44.000000000","message":"The whole point is to agree the mission so its clear on their importance.","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":5511,"name":"Nikola Dipanov","email":"ndipanov@redhat.com","username":"ndipanov"},"change_message_id":"49fa68011dde76b813041438626897132e6420d4","unresolved":false,"context_lines":[{"line_number":103,"context_line":"We do not develop any deployment or packaging for production deployments."},{"line_number":104,"context_line":"Being widely adopted by many distributions and commercial products, we"},{"line_number":105,"context_line":"instead choose to work with all those parties to ensure they are able to"},{"line_number":106,"context_line":"effectively package and deploy Nova."},{"line_number":107,"context_line":""},{"line_number":108,"context_line":"Scalability"},{"line_number":109,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_86ed93f1","line":106,"updated":"2015-05-01 14:32:12.000000000","message":"Very useful to clarify!","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":5511,"name":"Nikola Dipanov","email":"ndipanov@redhat.com","username":"ndipanov"},"change_message_id":"49fa68011dde76b813041438626897132e6420d4","unresolved":false,"context_lines":[{"line_number":132,"context_line":"There are some features that would not scale well for either the small scale"},{"line_number":133,"context_line":"or the very large scale. Ideally we should not accept these features, but if"},{"line_number":134,"context_line":"there is a strong case to add such features, we should work hard to ensure"},{"line_number":135,"context_line":"you can run without that feature at the scale you are required to run."},{"line_number":136,"context_line":""},{"line_number":137,"context_line":"Don\u0027t break exisiting users"},{"line_number":138,"context_line":"----------------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_b277a539","line":135,"updated":"2015-05-01 14:32:12.000000000","message":"This whole paragraph also does not make a lot of sense to me - we should not add features that will severly limit scalability, or don\u0027t work at scale, period.","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":5511,"name":"Nikola Dipanov","email":"ndipanov@redhat.com","username":"ndipanov"},"change_message_id":"49fa68011dde76b813041438626897132e6420d4","unresolved":false,"context_lines":[{"line_number":170,"context_line":""},{"line_number":171,"context_line":"The first example is the API that is now a proxy to the Glance v1 API."},{"line_number":172,"context_line":"As Glance moves to deprecate its v1 API, we now need to translate calls"},{"line_number":173,"context_line":"from the old v1 API we expose, to Glance\u0027s v2 API."},{"line_number":174,"context_line":""},{"line_number":175,"context_line":"The next API to mention is the networking APIs, in particular the"},{"line_number":176,"context_line":"security groups API. If you are using nova-network, Nova is still the only"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_bc853eae","line":173,"updated":"2015-05-01 14:32:12.000000000","message":"This is confusing to me - if it\u0027s being deprecated in Glance, why can\u0027t we also just deprecate ours?\n\nOn a side note - since this goes into the tree (not an email) and is built as docs (not necessarily only read from git) - we should use \"early-Liberty\" instead of \"now\"","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":5511,"name":"Nikola Dipanov","email":"ndipanov@redhat.com","username":"ndipanov"},"change_message_id":"49fa68011dde76b813041438626897132e6420d4","unresolved":false,"context_lines":[{"line_number":186,"context_line":""},{"line_number":187,"context_line":"A frequently raised topic is adding a Nova API to move all instances from"},{"line_number":188,"context_line":"one hypervisor to another, ideally automatically if we detect the hypervisor"},{"line_number":189,"context_line":"is known to be having problems."},{"line_number":190,"context_line":""},{"line_number":191,"context_line":"The Nova project has chosen to focus on providing a solid foundation for such"},{"line_number":192,"context_line":"functionality to be implemented by a system external to Nova."}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_f9850807","line":189,"updated":"2015-05-01 14:32:12.000000000","message":"This is an odd paragraph to have seeing that we already implement \"host_maintanance_mode\". I might be misunderstanding what this is getting at.\n\nWhile I fully agree that orchestration should not be something nova does and neither is monitoring, it may be a bit confusing that we are mentioning this specific case .","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"},{"author":{"_account_id":5511,"name":"Nikola Dipanov","email":"ndipanov@redhat.com","username":"ndipanov"},"change_message_id":"49fa68011dde76b813041438626897132e6420d4","unresolved":false,"context_lines":[{"line_number":204,"context_line":"to ensure we work well with the existing monitoring ecosystem."},{"line_number":205,"context_line":""},{"line_number":206,"context_line":"No native support for Batch Processing"},{"line_number":207,"context_line":"---------------------------------------"},{"line_number":208,"context_line":""},{"line_number":209,"context_line":"Currently Nova focuses on providing on demand compute resources in the style"},{"line_number":210,"context_line":"of classic Infrastructure as a Service clouds. A large pool of compute"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3acd31a7_dc78fac6","line":207,"updated":"2015-05-01 14:32:12.000000000","message":"I have to say that I don\u0027t understand what this whole section is about - seems like a glorified \u0027no\u0027 to a very specific feature request.","commit_id":"1a0fa26996f1fe4a8d1eb849fa4c69caf48c0d83"}]}