)]}'
{"doc/source/main/glossary.rst":[{"author":{"_account_id":10850,"name":"German Eichberger","email":"german.eichberger@gmail.com","username":"german"},"change_message_id":"2a34c72c3954469ec23fd357ce835208ca30dac2","unresolved":false,"context_lines":[{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"    Amphora"},{"line_number":16,"context_line":"        Virtual machine, container, dedicated hardware, appliance or device"},{"line_number":17,"context_line":"        that actually performs the task of load balancing in the Octavia"},{"line_number":18,"context_line":"        system. More specifically, an amphora takes requests from clients on"},{"line_number":19,"context_line":"        the front-end and distributes these to back-end systems. Amphorae"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa98f980_6c84890b","line":16,"updated":"2014-09-18 21:07:51.000000000","message":"dedicated hardware - this is ambiguous: This means I could manage my Netscaler with Octavia, too, or does this just meen bare metal VMs?","commit_id":"253c7cded6476c6fd2ac9d276fe855380046cf79"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"7247b302cf6ab2fb17067bcac4c3e38b8ed168e0","unresolved":false,"context_lines":[{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"    Amphora"},{"line_number":16,"context_line":"        Virtual machine, container, dedicated hardware, appliance or device"},{"line_number":17,"context_line":"        that actually performs the task of load balancing in the Octavia"},{"line_number":18,"context_line":"        system. More specifically, an amphora takes requests from clients on"},{"line_number":19,"context_line":"        the front-end and distributes these to back-end systems. Amphorae"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa98f980_f8b75038","line":16,"in_reply_to":"fa98f980_6c84890b","updated":"2014-09-19 02:19:15.000000000","message":"If your Netscaler is capable of spinning up virtual load balancing service instances that operate in a way compatible with the Octavia networking and control model (ie. as something compatible enough with our reference implementation amphorae), then I\u0027m not going to rule it out. I suspect most Netscalers won\u0027t be able to do this unless the vendor alters their product specifically to do this. But I\u0027m also not making any commitments about what will or will not work with Octavia here-- that\u0027s what the design documents and code are for.\n\nIs the definition of what an amphora is unclear here? (That is to say, would someone unfamiliar with Octavia not know what an amphora is after reading this definition?)","commit_id":"253c7cded6476c6fd2ac9d276fe855380046cf79"},{"author":{"_account_id":10850,"name":"German Eichberger","email":"german.eichberger@gmail.com","username":"german"},"change_message_id":"2a34c72c3954469ec23fd357ce835208ca30dac2","unresolved":false,"context_lines":[{"line_number":31,"context_line":"    Apolocation"},{"line_number":32,"context_line":"        Term used to describe when two or more amphorae are not colocated on"},{"line_number":33,"context_line":"        the same physical hardware (which is often essential in HA topologies)."},{"line_number":34,"context_line":"        May also be used to describe two or more loadbalancers which are not"},{"line_number":35,"context_line":"        colocated on the same amphora."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"    Controller"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa98f980_0c8e9d27","line":34,"updated":"2014-09-18 21:07:51.000000000","message":"I would stick to the first definiton and not even entertain the second.","commit_id":"253c7cded6476c6fd2ac9d276fe855380046cf79"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"7247b302cf6ab2fb17067bcac4c3e38b8ed168e0","unresolved":false,"context_lines":[{"line_number":31,"context_line":"    Apolocation"},{"line_number":32,"context_line":"        Term used to describe when two or more amphorae are not colocated on"},{"line_number":33,"context_line":"        the same physical hardware (which is often essential in HA topologies)."},{"line_number":34,"context_line":"        May also be used to describe two or more loadbalancers which are not"},{"line_number":35,"context_line":"        colocated on the same amphora."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"    Controller"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa98f980_787b607b","line":34,"in_reply_to":"fa98f980_0c8e9d27","updated":"2014-09-19 02:19:15.000000000","message":"We need something to describe the latter as well since it will come up in conversation. In its most generic sense \u0027apolocation\u0027 is the opposite of \u0027colocation.\u0027 Apolocation simply describes the condition of not being together. Should I split this definition up into two separate terms (one for apolocation of amphorae, and one for apolocation of load balancers)? Would this help to avoid confusion?","commit_id":"253c7cded6476c6fd2ac9d276fe855380046cf79"},{"author":{"_account_id":10850,"name":"German Eichberger","email":"german.eichberger@gmail.com","username":"german"},"change_message_id":"2a34c72c3954469ec23fd357ce835208ca30dac2","unresolved":false,"context_lines":[{"line_number":35,"context_line":"        colocated on the same amphora."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"    Controller"},{"line_number":38,"context_line":"        Daemon with access to both the LB Network and OpenStack components"},{"line_number":39,"context_line":"        which coordinates and manages the overall activity of the Octavia load"},{"line_number":40,"context_line":"        balancing system. Controllers will usually use an abstracted driver"},{"line_number":41,"context_line":"        interface (usually a base class) for communicating with various other"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa98f980_6c9ba966","line":38,"updated":"2014-09-18 21:07:51.000000000","message":"Daemon \u003d implementation detail. For all we lknow we might have more than one daemon, etc.","commit_id":"253c7cded6476c6fd2ac9d276fe855380046cf79"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"7247b302cf6ab2fb17067bcac4c3e38b8ed168e0","unresolved":false,"context_lines":[{"line_number":35,"context_line":"        colocated on the same amphora."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"    Controller"},{"line_number":38,"context_line":"        Daemon with access to both the LB Network and OpenStack components"},{"line_number":39,"context_line":"        which coordinates and manages the overall activity of the Octavia load"},{"line_number":40,"context_line":"        balancing system. Controllers will usually use an abstracted driver"},{"line_number":41,"context_line":"        interface (usually a base class) for communicating with various other"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa98f980_383638b5","line":38,"in_reply_to":"fa98f980_6c9ba966","updated":"2014-09-19 02:19:15.000000000","message":"We are going to need to be able to discuss implementation details because Octavia is an implementation, and a controller will necessarily have a hell of a lot of said details. If a controller isn\u0027t a daemon, what else is it?\n\nPlease note that I\u0027m defining what \"a controller\" is, not what \"controllers\" are and I\u0027m making no indication here that there will be only one of them. (This is the same treatment above with \"Amphora,\" and I\u0027m certainly not implying above that there will be only one amphora.) In fact, later in this definition I refer to \"controllers\" in the plural, which implies there can be more than one.\n\nHaving said this, in version 0.5 there will be only one controller. (Multiple controllers happen with version 1.) However, we will need to separate this term into \"Active Amphorae Controller\" and \"Spare Amphorae Controller\" when we move to version 1. Do you think it\u0027s worthwhile to define these separately right now?","commit_id":"253c7cded6476c6fd2ac9d276fe855380046cf79"},{"author":{"_account_id":12403,"name":"Al Miller","email":"al.miller@ajmiller.net","username":"ajmiller"},"change_message_id":"3a1fe81249eb9152094eff860ba92959eb1288de","unresolved":false,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Note also that many of these terms are expanded upon in design documents in"},{"line_number":10,"context_line":"this same repository. What follows is a brief but necessarily incomplete"},{"line_number":11,"context_line":"description of these terms."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_a11dcabf","line":11,"updated":"2014-09-22 16:52:04.000000000","message":"I realize that this glossary needs to be short and concise, but it uses lots of \"soft\" words like \"may\", \"might\", \"usually\" that leave lots of room for interpretation and potential confusion.  They may be appropriate, but it would be good to look at these and replace them with more definitive words when appropriate.","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"ceffcabd2bb5c5849ed8660f7f88821c173d23da","unresolved":false,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Note also that many of these terms are expanded upon in design documents in"},{"line_number":10,"context_line":"this same repository. What follows is a brief but necessarily incomplete"},{"line_number":11,"context_line":"description of these terms."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_f58f7e21","line":11,"in_reply_to":"da9df570_a11dcabf","updated":"2014-09-23 05:00:46.000000000","message":"Hmmm... reading through... when the language is less precise, it\u0027s usually that way for a reason. Could you point out some specific cases where you think we should use harder language?  (Or rather, are there any ones other than the few you did point out that you\u0027d like to see changed?)","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":12403,"name":"Al Miller","email":"al.miller@ajmiller.net","username":"ajmiller"},"change_message_id":"3a1fe81249eb9152094eff860ba92959eb1288de","unresolved":false,"context_lines":[{"line_number":72,"context_line":""},{"line_number":73,"context_line":"    Load Balancer"},{"line_number":74,"context_line":"        Object describing a logical grouping of listeners on one or more VIPs"},{"line_number":75,"context_line":"        and associated with one or more amphorae. (Our \"Loadbalancer\" most"},{"line_number":76,"context_line":"        closely resembles a Virtual IP address in other load balancing"},{"line_number":77,"context_line":"        implementations.) Whether the load balancer exists on more than one"},{"line_number":78,"context_line":"        amphora depends on the topology used. The load balancer is also often"}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_3654c253","line":75,"updated":"2014-09-22 16:52:04.000000000","message":"For consistency: \"Loadbalancer\" -\u003e \"Load Balancer\"","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"ceffcabd2bb5c5849ed8660f7f88821c173d23da","unresolved":false,"context_lines":[{"line_number":72,"context_line":""},{"line_number":73,"context_line":"    Load Balancer"},{"line_number":74,"context_line":"        Object describing a logical grouping of listeners on one or more VIPs"},{"line_number":75,"context_line":"        and associated with one or more amphorae. (Our \"Loadbalancer\" most"},{"line_number":76,"context_line":"        closely resembles a Virtual IP address in other load balancing"},{"line_number":77,"context_line":"        implementations.) Whether the load balancer exists on more than one"},{"line_number":78,"context_line":"        amphora depends on the topology used. The load balancer is also often"}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_15bc02bb","line":75,"in_reply_to":"da9df570_3654c253","updated":"2014-09-23 05:00:46.000000000","message":"Right: So proper English would say we should use \"Load Balancer.\" But the name of the object is actually \"Loadbalancer.\" I could replace all the former with the latter, eh. Do you think I should?","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":12403,"name":"Al Miller","email":"al.miller@ajmiller.net","username":"ajmiller"},"change_message_id":"3a1fe81249eb9152094eff860ba92959eb1288de","unresolved":false,"context_lines":[{"line_number":88,"context_line":""},{"line_number":89,"context_line":"    Member"},{"line_number":90,"context_line":"        Object representing a single back-end server or system that is a"},{"line_number":91,"context_line":"        part of a pool. A member may be associated with only one pool."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"    Pool"},{"line_number":94,"context_line":"        Object representing the grouping of members to which the listener"}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_dc52b512","line":91,"updated":"2014-09-22 16:52:04.000000000","message":"Can a member every be associated with more that one pool?  If not, then \"may\" -\u003e \"will\" might be better.  If so, then a brief explanation about when this can happen would be helpful.","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"ceffcabd2bb5c5849ed8660f7f88821c173d23da","unresolved":false,"context_lines":[{"line_number":88,"context_line":""},{"line_number":89,"context_line":"    Member"},{"line_number":90,"context_line":"        Object representing a single back-end server or system that is a"},{"line_number":91,"context_line":"        part of a pool. A member may be associated with only one pool."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"    Pool"},{"line_number":94,"context_line":"        Object representing the grouping of members to which the listener"}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_b5c41641","line":91,"in_reply_to":"da9df570_dc52b512","updated":"2014-09-23 05:00:46.000000000","message":"Fixing. (Gonna use \"is\" instead of \"may be\")","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":12403,"name":"Al Miller","email":"al.miller@ajmiller.net","username":"ajmiller"},"change_message_id":"3a1fe81249eb9152094eff860ba92959eb1288de","unresolved":false,"context_lines":[{"line_number":94,"context_line":"        Object representing the grouping of members to which the listener"},{"line_number":95,"context_line":"        forwards client requests. Note that a pool may be associated with only"},{"line_number":96,"context_line":"        one listener, but a listener might refer to several pools (and switch"},{"line_number":97,"context_line":"        between them using layer-7 rules)."},{"line_number":98,"context_line":""},{"line_number":99,"context_line":"    VIP"},{"line_number":100,"context_line":"    Virtual IP Address"}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_a1c5ca5c","line":97,"updated":"2014-09-22 16:52:04.000000000","message":"The \"may\" and \"might\" here make this note sound confusing and weak.   Can a pool ever be associated with more than one listener?  Would a better wording be something like \"Note that while a pool is associated with only one listener, a listener might refer to several pools (and switch between them using layer-7 rules).\"","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"ceffcabd2bb5c5849ed8660f7f88821c173d23da","unresolved":false,"context_lines":[{"line_number":94,"context_line":"        Object representing the grouping of members to which the listener"},{"line_number":95,"context_line":"        forwards client requests. Note that a pool may be associated with only"},{"line_number":96,"context_line":"        one listener, but a listener might refer to several pools (and switch"},{"line_number":97,"context_line":"        between them using layer-7 rules)."},{"line_number":98,"context_line":""},{"line_number":99,"context_line":"    VIP"},{"line_number":100,"context_line":"    Virtual IP Address"}],"source_content_type":"text/x-rst","patch_set":3,"id":"da9df570_35ff4669","line":97,"in_reply_to":"da9df570_a1c5ca5c","updated":"2014-09-23 05:00:46.000000000","message":"Done","commit_id":"8ae0c7f95a08699cbf71450ca615189692e452bf"},{"author":{"_account_id":6951,"name":"Brandon Logan","email":"brandon.logan@rackspace.com","username":"brandon-logan"},"change_message_id":"54803a98d934195ba627a73df6eaa070a0f2589d","unresolved":false,"context_lines":[{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"    Amphora"},{"line_number":16,"context_line":"        Virtual machine, container, dedicated hardware, appliance or device"},{"line_number":17,"context_line":"        that actually performs the task of load balancing in the Octavia"},{"line_number":18,"context_line":"        system. More specifically, an amphora takes requests from clients on"},{"line_number":19,"context_line":"        the front-end and distributes these to back-end systems. Amphorae"}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_35a02643","line":16,"updated":"2014-09-23 05:26:33.000000000","message":"funny how you call them appliance or device.  If only some brave soul had championed those two names. oh wait...\n\nI will never forget!","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"34c3573b316df06b9d5ba528e1b074193d37cbaf","unresolved":false,"context_lines":[{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"    Amphora"},{"line_number":16,"context_line":"        Virtual machine, container, dedicated hardware, appliance or device"},{"line_number":17,"context_line":"        that actually performs the task of load balancing in the Octavia"},{"line_number":18,"context_line":"        system. More specifically, an amphora takes requests from clients on"},{"line_number":19,"context_line":"        the front-end and distributes these to back-end systems. Amphorae"}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_10889070","line":16,"in_reply_to":"da9df570_35a02643","updated":"2014-09-23 05:34:17.000000000","message":"It\u0027s the gift that keeps on giving!","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":6951,"name":"Brandon Logan","email":"brandon.logan@rackspace.com","username":"brandon-logan"},"change_message_id":"54803a98d934195ba627a73df6eaa070a0f2589d","unresolved":false,"context_lines":[{"line_number":20,"context_line":"        communicate with their controllers over the LB Network through a driver"},{"line_number":21,"context_line":"        interface on the controller."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"    Amphora Driver"},{"line_number":24,"context_line":"        Component of the controller that does all the communication with"},{"line_number":25,"context_line":"        amphorae. Drivers communicate with the controller through a generic"},{"line_number":26,"context_line":"        base class and associated methods, and translate these into control"}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_b59336f9","line":23,"updated":"2014-09-23 05:26:33.000000000","message":"Change this to Amphora Load Balancer Driver...I believe we agreed to this in a vote.","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"34c3573b316df06b9d5ba528e1b074193d37cbaf","unresolved":false,"context_lines":[{"line_number":20,"context_line":"        communicate with their controllers over the LB Network through a driver"},{"line_number":21,"context_line":"        interface on the controller."},{"line_number":22,"context_line":""},{"line_number":23,"context_line":"    Amphora Driver"},{"line_number":24,"context_line":"        Component of the controller that does all the communication with"},{"line_number":25,"context_line":"        amphorae. Drivers communicate with the controller through a generic"},{"line_number":26,"context_line":"        base class and associated methods, and translate these into control"}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_70963c96","line":23,"in_reply_to":"da9df570_b59336f9","updated":"2014-09-23 05:34:17.000000000","message":"Aah! You\u0027re right. Fixing.","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"9a67e7ea3b47b71d95de05dd5df5f8031d485514","unresolved":false,"context_lines":[{"line_number":32,"context_line":"        Term used to describe when two or more amphorae are not colocated on"},{"line_number":33,"context_line":"        the same physical hardware (which is often essential in HA topologies)."},{"line_number":34,"context_line":"        May also be used to describe two or more loadbalancers which are not"},{"line_number":35,"context_line":"        colocated on the same amphora."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"    Controller"},{"line_number":38,"context_line":"        Daemon with access to both the LB Network and OpenStack components"}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_35e90681","line":35,"updated":"2014-09-23 05:10:28.000000000","message":"I don\u0027t want to lose German\u0027s point on this term from PS 2: Do y\u0027all think I should be specific about the difference between \u0027load balancer apolocation\u0027 and \u0027amphora apolocation\u0027? Or is just defining the term \u0027apolocation\u0027 enough?","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":6951,"name":"Brandon Logan","email":"brandon.logan@rackspace.com","username":"brandon-logan"},"change_message_id":"2168609f6d5472855d53a6af10e3f6ab9d5d13ed","unresolved":false,"context_lines":[{"line_number":32,"context_line":"        Term used to describe when two or more amphorae are not colocated on"},{"line_number":33,"context_line":"        the same physical hardware (which is often essential in HA topologies)."},{"line_number":34,"context_line":"        May also be used to describe two or more loadbalancers which are not"},{"line_number":35,"context_line":"        colocated on the same amphora."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"    Controller"},{"line_number":38,"context_line":"        Daemon with access to both the LB Network and OpenStack components"}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_d05fe80c","line":35,"in_reply_to":"da9df570_35e90681","updated":"2014-09-23 05:30:12.000000000","message":"I think in conversation we will just specify if what granularity of apolocation we are talking about.  Or by default we will assume it is physical hardware apolocation and specifically say amphora apolocation for hte second.","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":6951,"name":"Brandon Logan","email":"brandon.logan@rackspace.com","username":"brandon-logan"},"change_message_id":"54803a98d934195ba627a73df6eaa070a0f2589d","unresolved":false,"context_lines":[{"line_number":106,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":107,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":108,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":109,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_302114a0","line":109,"updated":"2014-09-23 05:26:33.000000000","message":"Maybe Add:\n\nTLS Termination\n\nL7 Rules/Switching","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"34c3573b316df06b9d5ba528e1b074193d37cbaf","unresolved":false,"context_lines":[{"line_number":106,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":107,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":108,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":109,"context_line":""}],"source_content_type":"text/x-rst","patch_set":4,"id":"da9df570_b0e7c43f","line":109,"in_reply_to":"da9df570_302114a0","updated":"2014-09-23 05:34:17.000000000","message":"Will do!","commit_id":"41ccce9df1d3d8233705c7c8d30e5e8a30115d6a"},{"author":{"_account_id":11628,"name":"Michael Johnson","email":"johnsomor@gmail.com","username":"johnsom"},"change_message_id":"233b6bc6b54d24da11309217036ca3157ed11edf","unresolved":false,"context_lines":[{"line_number":10,"context_line":"this same repository. What follows is a brief but necessarily incomplete"},{"line_number":11,"context_line":"description of these terms."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"    Amphora"},{"line_number":16,"context_line":"        Virtual machine, container, dedicated hardware, appliance or device"}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_1dbded68","line":13,"updated":"2014-09-23 20:51:32.000000000","message":"Is glossary part of reStructuredText?  I can\u0027t find it on the docutils directives reference page: http://docutils.sourceforge.net/docs/ref/rst/directives.html and my review tool (ReText) throws an error on it.","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"4f5550863849a6a5ed5b00f07c2be847ec940d1a","unresolved":false,"context_lines":[{"line_number":10,"context_line":"this same repository. What follows is a brief but necessarily incomplete"},{"line_number":11,"context_line":"description of these terms."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":".. glossary:: :sorted:"},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"    Amphora"},{"line_number":16,"context_line":"        Virtual machine, container, dedicated hardware, appliance or device"}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_cbfc7b41","line":13,"in_reply_to":"da9df570_1dbded68","updated":"2014-09-23 21:54:37.000000000","message":"Yes, it is.  http://sphinx-doc.org/markup/para.html\n\nAnd in any case, the documents generated from this are formatted like a glossary properly.","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11628,"name":"Michael Johnson","email":"johnsomor@gmail.com","username":"johnsom"},"change_message_id":"233b6bc6b54d24da11309217036ca3157ed11edf","unresolved":false,"context_lines":[{"line_number":80,"context_line":""},{"line_number":81,"context_line":"    L7 Switching"},{"line_number":82,"context_line":"    Layer 7 Switching"},{"line_number":83,"context_line":"        This is a load balancing feature specific to HTTP or terminated HTTPS"},{"line_number":84,"context_line":"        sessions, in which different client requests are routed to different"},{"line_number":85,"context_line":"        back-end pools depending on one or more layer 7 policies the user might"},{"line_number":86,"context_line":"        configure."}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_600ace22","line":83,"updated":"2014-09-23 20:51:32.000000000","message":"You can layer 7 switch protocols other than HTTP such as VPN connections, etc.  It\u0027s a nitpick, but it may save a change to this later.","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"4f5550863849a6a5ed5b00f07c2be847ec940d1a","unresolved":false,"context_lines":[{"line_number":80,"context_line":""},{"line_number":81,"context_line":"    L7 Switching"},{"line_number":82,"context_line":"    Layer 7 Switching"},{"line_number":83,"context_line":"        This is a load balancing feature specific to HTTP or terminated HTTPS"},{"line_number":84,"context_line":"        sessions, in which different client requests are routed to different"},{"line_number":85,"context_line":"        back-end pools depending on one or more layer 7 policies the user might"},{"line_number":86,"context_line":"        configure."}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_8bb0b38c","line":83,"in_reply_to":"da9df570_600ace22","updated":"2014-09-23 21:54:37.000000000","message":"Except that we don\u0027t in either Neutron LBaaS or Octavia. Layer 7 stuff in current implementations applies only to HTTP or terminated HTTP protocol connections. I\u0027d rather stick with this definition and change it later (or introduce a near term) if we ever decide to support other protocols for this type of switching.","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11628,"name":"Michael Johnson","email":"johnsomor@gmail.com","username":"johnsom"},"change_message_id":"233b6bc6b54d24da11309217036ca3157ed11edf","unresolved":false,"context_lines":[{"line_number":123,"context_line":"    Member"},{"line_number":124,"context_line":"        Object representing a single back-end server or system that is a"},{"line_number":125,"context_line":"        part of a pool. A member is associated with only one pool."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"    Pool"},{"line_number":128,"context_line":"        Object representing the grouping of members to which the listener"},{"line_number":129,"context_line":"        forwards client requests. Note that a pool is associated with only"}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_0073c215","line":126,"updated":"2014-09-23 20:51:32.000000000","message":"We are missing \"Octavia\"\nNewbies get Octavia and Amphora confused","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"4f5550863849a6a5ed5b00f07c2be847ec940d1a","unresolved":false,"context_lines":[{"line_number":123,"context_line":"    Member"},{"line_number":124,"context_line":"        Object representing a single back-end server or system that is a"},{"line_number":125,"context_line":"        part of a pool. A member is associated with only one pool."},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"    Pool"},{"line_number":128,"context_line":"        Object representing the grouping of members to which the listener"},{"line_number":129,"context_line":"        forwards client requests. Note that a pool is associated with only"}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_eb7cdfb7","line":126,"in_reply_to":"da9df570_0073c215","updated":"2014-09-23 21:54:37.000000000","message":"Aah-- good catch. I\u0027ll add it!","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11628,"name":"Michael Johnson","email":"johnsomor@gmail.com","username":"johnsom"},"change_message_id":"233b6bc6b54d24da11309217036ca3157ed11edf","unresolved":false,"context_lines":[{"line_number":149,"context_line":"        http://en.wikipedia.org/wiki/Virtual_IP_address In other"},{"line_number":150,"context_line":"        load balancing implementations, this might be called the \"floating IP.\""},{"line_number":151,"context_line":"        (Though it is important to note this is not necessarily a Neutron"},{"line_number":152,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":153,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":154,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":155,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_c0c9ba7a","line":152,"updated":"2014-09-23 20:51:32.000000000","message":"It might be beneficial to include a definition of Neutron Floating IP (or just Floating IPs) and call out that VIPs may or may not be allocated using Neutron Floating IPs.\nIt\u0027s a bit confusing because VRRP, CARP, and HSRP call these VIPs.\nMaybe even drop \"floating IP\" from this section....","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"4f5550863849a6a5ed5b00f07c2be847ec940d1a","unresolved":false,"context_lines":[{"line_number":149,"context_line":"        http://en.wikipedia.org/wiki/Virtual_IP_address In other"},{"line_number":150,"context_line":"        load balancing implementations, this might be called the \"floating IP.\""},{"line_number":151,"context_line":"        (Though it is important to note this is not necessarily a Neutron"},{"line_number":152,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":153,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":154,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":155,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_e6b41637","line":152,"in_reply_to":"da9df570_c0c9ba7a","updated":"2014-09-23 21:54:37.000000000","message":"So, a neutron floating IP is really just a static NAT, whereas in pretty much every other context in the IT industry a \"Floating IP\" is a VIP as described in the wikipedia link referenced above. Yet another instance where Neutron decides to mis-use terms and therefore inject a healthy dose of confusion into the process. :P\n\nI felt it was a good idea to mention \"floating IP\" so that those familiar with this term\u0027s usage in other contexts would know what we\u0027re talking about (and yet distinguish this from Neutron floating IPs in case people are also familiar with that mis-usage of the therm). But I can see how this does only make the definition more confusing-- and really doesn\u0027t add anything to the meaning beyond what is referenced in the VIP wikipedia article, so I\u0027ll remove mention of it here.","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":10850,"name":"German Eichberger","email":"german.eichberger@gmail.com","username":"german"},"change_message_id":"97bb2b0df9f47a4bb442a1881bf0020c04252c34","unresolved":false,"context_lines":[{"line_number":151,"context_line":"        (Though it is important to note this is not necessarily a Neutron"},{"line_number":152,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":153,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":154,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":155,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_a02596c3","line":154,"updated":"2014-09-23 20:59:03.000000000","message":"Are those supported by Neutron?I only found a blueprint supporting VRRP (see https://blueprints.launchpad.net/neutron/+spec/l3-high-availability). \n\nI would suggest  not describing things which won\u0027t work in Neutron or don\u0027t have blueprints --\n\n I only did a limited search (see https://www.openstack.org/assets/presentation-media/L3-Agent-HA.pdf) and that ARPing takes 60 s was worrying enough ;-)","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":10850,"name":"German Eichberger","email":"german.eichberger@gmail.com","username":"german"},"change_message_id":"a298136c618b9fd77144b39cd7465b3d106977ec","unresolved":false,"context_lines":[{"line_number":151,"context_line":"        (Though it is important to note this is not necessarily a Neutron"},{"line_number":152,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":153,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":154,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":155,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_eb53ff36","line":154,"in_reply_to":"da9df570_a02596c3","updated":"2014-09-23 21:30:42.000000000","message":"Scratch that... it seems that might work...","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"},{"author":{"_account_id":11685,"name":"Stephen Balukoff","email":"stephen@balukoff.com","username":"sbalukoff"},"change_message_id":"4f5550863849a6a5ed5b00f07c2be847ec940d1a","unresolved":false,"context_lines":[{"line_number":151,"context_line":"        (Though it is important to note this is not necessarily a Neutron"},{"line_number":152,"context_line":"        floating IP.) In a highly available load balancing topology in Octavia,"},{"line_number":153,"context_line":"        the VIP might be assigned to several amphorae, and a layer-2 protocol"},{"line_number":154,"context_line":"        like CARP, VRRP, or HSRP might be used to maintain its availability."},{"line_number":155,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"da9df570_465962fd","line":154,"in_reply_to":"da9df570_a02596c3","updated":"2014-09-23 21:54:37.000000000","message":"You might have noticed how very few of the other definitions in here make reference to Neutron-specific technologies and terminology. This is because the hope is to keep Octavia loosely coupled with the networking layer. This is also the reason we have a networking driver which will interface with Neutron.\n\nIt\u0027s true that the first (and perhaps only) driver we\u0027ll write will be to interface with Neutron. But if someone wants to write a networking driver to interface with nova-networking or some 3rd party SDN system, I don\u0027t want to be so tied to Neutron that they can\u0027t do that. In other networking systems, other layer-2 high availability protocols might be usable.\n\nAnd in any case, the point here is to give the reader of this glossary a solid understanding of what we mean by VIP (especially if they\u0027re coming from the Neutron LBaaS v1 world, where the term, again, is horribly mis-used.) We\u0027re not saying, in this definition, that the particular networking layer in use by Octavia must support any one of the protocols mentioned.\n\nAnyway-- I also notice I forgot to mention how VIPs might be used in routed topologies, so I\u0027ll add a sentence clarifying that here.","commit_id":"ffaf35e315dccdc7b96cfb73af66a7809f79d158"}]}
