)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"c242c33ae44cc59333eab32a9f0f27a488848fea","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Deprecate segment_mtu and add global_physnet_mtu"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This deprecate\u0027s the ML2 specific \u0027segment_mtu\u0027 option and"},{"line_number":10,"context_line":"replaces it with Neutron-wide \u0027global_physnet_mtu\u0027 option"},{"line_number":11,"context_line":"that should be used by all plugins to specify the default"},{"line_number":12,"context_line":"MTU for networks."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"9aed3d3a_c2722ee5","line":9,"updated":"2016-02-26 12:32:30.000000000","message":"It\u0027s my belief that configuration options should not serve as part of plugin API. Instead, we should provide a function to access the option (get_global_physnet_mtu), and use it in plugins that are interested in it.\n\nThat way, we free ourselves to change the option name or even the implementation of the extraction of the value later without breaking any plugin.","commit_id":"150e73c7441d46eda2954140e2964a955c3b1e55"},{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"a263a9edcfb62f448a86eea956b4ae2cfef55f00","unresolved":false,"context_lines":[{"line_number":9,"context_line":"This deprecate\u0027s the ML2 specific \u0027segment_mtu\u0027 option and"},{"line_number":10,"context_line":"replaces it with Neutron-wide \u0027global_physnet_mtu\u0027 option"},{"line_number":11,"context_line":"that should be used by all plugins to specify the default"},{"line_number":12,"context_line":"MTU for networks."},{"line_number":13,"context_line":""},{"line_number":14,"context_line":"Change-Id: I6ffc8973c9b8f46cc19922ff04fdd2d23646b878"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"5aef4532_4cbad024","line":12,"updated":"2016-03-08 04:30:30.000000000","message":"a bug would be nice. Was this release noted?","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"3ad20a9a758b9d1d4272b0a4dd00454b0f5a790a","unresolved":false,"context_lines":[{"line_number":9,"context_line":"Introduce the neutron-wide \u0027global_physnet_mtu\u0027 option that"},{"line_number":10,"context_line":"references the underlying physical network MTU. All plug-ins"},{"line_number":11,"context_line":"should use this option to calculate the proper MTU for virtual"},{"line_number":12,"context_line":"network components. Also deprecate the \u0027segment_mtu\u0027 option"},{"line_number":13,"context_line":"specific to the ML2 plug-in."},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"Closes-Bug: #1542475"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":9,"id":"5aef4532_5f508f5d","line":12,"updated":"2016-03-08 19:05:01.000000000","message":"OK, it\u0027s my time to be pain here: the commit message should mention the wrapper function and suggest all plugins should rely on it instead of the option. [the current wording suggests they should access CONF directly]","commit_id":"3f376e884434bbc2b6e990052f5a76b9df80498e"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"e5f305b47dfc0bdf8674e95bfa7d4b028509899c","unresolved":false,"context_lines":[{"line_number":9,"context_line":"Introduce the neutron-wide \u0027global_physnet_mtu\u0027 option that"},{"line_number":10,"context_line":"references the underlying physical network MTU. All plug-ins"},{"line_number":11,"context_line":"should use this option to calculate the proper MTU for virtual"},{"line_number":12,"context_line":"network components. Also deprecate the \u0027segment_mtu\u0027 option"},{"line_number":13,"context_line":"specific to the ML2 plug-in."},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"Closes-Bug: #1542475"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":9,"id":"5aef4532_c25b242d","line":12,"in_reply_to":"5aef4532_5f508f5d","updated":"2016-03-08 19:11:16.000000000","message":"Done","commit_id":"3f376e884434bbc2b6e990052f5a76b9df80498e"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"2580f97707719895af3a9513e8415aa53527898d","unresolved":false,"context_lines":[{"line_number":12,"context_line":"should use to retrieve it. This value should be used to"},{"line_number":13,"context_line":"calculate the proper MTU for virtual network components."},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"This patch also deprecate the \u0027segment_mtu\u0027 option specific"},{"line_number":16,"context_line":"to the ML2 plug-in and makes ML2 reference this new option."},{"line_number":17,"context_line":""},{"line_number":18,"context_line":"Closes-Bug: #1542475"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":10,"id":"5aef4532_a2d1b859","line":15,"range":{"start_line":15,"start_character":16,"end_line":15,"end_character":25},"updated":"2016-03-08 19:12:18.000000000","message":"deprecates","commit_id":"ae45cd5732b8b75a5b2d35df555f46e11145eb85"}],"neutron/common/config.py":[{"author":{"_account_id":8124,"name":"cbrandily","email":"zzelle@gmail.com","username":"cbrandily"},"change_message_id":"24086011895e6192c9433d0208dcf3adf19c6969","unresolved":false,"context_lines":[{"line_number":185,"context_line":"# NOTE(kevinbenton): this is in the main config because the global_physnet_mtu"},{"line_number":186,"context_line":"# default value depends on it as part of the segment_mtu deprecation."},{"line_number":187,"context_line":"ml2_opts \u003d ["},{"line_number":188,"context_line":"    cfg.IntOpt(\u0027segment_mtu\u0027, default\u003d1500, deprecated_for_removal\u003dTrue,"},{"line_number":189,"context_line":"               help\u003d_(\u0027The maximum permissible size of an unfragmented \u0027"},{"line_number":190,"context_line":"                      \u0027packet travelling a L2 network segment. The default \u0027"},{"line_number":191,"context_line":"                      \u0027value of 1500 is used as the segment MTU, to reflect \u0027"}],"source_content_type":"text/x-python","patch_set":1,"id":"9aed3d3a_ed030c75","line":188,"updated":"2016-02-25 21:15:42.000000000","message":"Why not defining segment_mtu as a deprecated version of global_physnet_mtu? it avoids invalid config if someone provides both:\n\n conf.IntOpt(\u0027global_physnet_mtu\u0027,\n             deprecated_name\u003d\u0027segment_mtu\u0027,\n             deprecated_group\u003d\u0027ml2\u0027,","commit_id":"9353aadb0e582a59b051f2f9a650683aa7994fc2"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"27b06b5898580661c1b67d324b7a03d9c415f630","unresolved":false,"context_lines":[{"line_number":185,"context_line":"# NOTE(kevinbenton): this is in the main config because the global_physnet_mtu"},{"line_number":186,"context_line":"# default value depends on it as part of the segment_mtu deprecation."},{"line_number":187,"context_line":"ml2_opts \u003d ["},{"line_number":188,"context_line":"    cfg.IntOpt(\u0027segment_mtu\u0027, default\u003d1500, deprecated_for_removal\u003dTrue,"},{"line_number":189,"context_line":"               help\u003d_(\u0027The maximum permissible size of an unfragmented \u0027"},{"line_number":190,"context_line":"                      \u0027packet travelling a L2 network segment. The default \u0027"},{"line_number":191,"context_line":"                      \u0027value of 1500 is used as the segment MTU, to reflect \u0027"}],"source_content_type":"text/x-python","patch_set":1,"id":"9aed3d3a_937415d0","line":188,"in_reply_to":"9aed3d3a_ed030c75","updated":"2016-02-26 02:16:09.000000000","message":"Cool. I didn\u0027t realize that was an option.","commit_id":"9353aadb0e582a59b051f2f9a650683aa7994fc2"},{"author":{"_account_id":8873,"name":"Assaf Muller","email":"amuller@redhat.com","username":"amuller"},"change_message_id":"ecb63693e51c047c3fb55389941314da60b0d3ab","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_69b6c08b","line":179,"updated":"2016-03-01 16:44:06.000000000","message":"Are there plans to auto calculate this value instead of requiring the deployment tool to set it?","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":8873,"name":"Assaf Muller","email":"amuller@redhat.com","username":"amuller"},"change_message_id":"287504a45bef3413f1c5a700ea515f4bdb3fe2df","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_c28d297d","line":179,"in_reply_to":"7af24918_02527113","updated":"2016-03-01 17:08:59.000000000","message":"I didn\u0027t -1, I asked a question.","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"a4cd2ca9581284b080d1d0962938e533a034c440","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_84a678bf","line":179,"in_reply_to":"7af24918_69b6c08b","updated":"2016-03-01 23:50:30.000000000","message":"This is actually something quite difficult to do. Different interfaces on different compute nodes with different physnet_mappings can all have different MTUs. Picking the correct one could be quite hard.\n\nSimilarly, we wouldn\u0027t want to bump an entire jumbo frame-based deployment from 9000 down to 1500 suddenly because one compute node came online that had a misconfigured interface.\n\nI would be for it, but it\u0027s going to be quite an undertaking with lots of assumptions involved.\n\nI don\u0027t want to add a TODO here because that\u0027s something that would need to go through an RFE to see if it\u0027s something we can get everyone to agree is something we want to do.","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"f6883e096b8059eb67aee72bb75c2460cdbf7967","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_02527113","line":179,"in_reply_to":"7af24918_69b6c08b","updated":"2016-03-01 17:07:11.000000000","message":"While that\u0027s a noble goal, I think it\u0027s a separate step from enabling other plugins to use the option value, and hence should not block it. I am supportive of adding a note or TODO for the autocalculation feature though.","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":8873,"name":"Assaf Muller","email":"amuller@redhat.com","username":"amuller"},"change_message_id":"adb689865665ebf7b7b25db4771f90e477e57bad","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_a460d436","line":179,"in_reply_to":"7af24918_84a678bf","updated":"2016-03-01 23:56:06.000000000","message":"Yeah, I had similar thoughts after I left my comment, I just wanted to float the idea. My concern is that we introduce segment_mtu one cycle, rename it the next, then there\u0027s a scenario where we remove this option the following cycle. How deployment tools are supposed to keep up is beyond me.","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":9515,"name":"Matthew Kassawara","email":"mkassawara@gmail.com","username":"ionosphere80"},"change_message_id":"a74fa6c4e12b5cd109bdf32f35d5f3b1002015be","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_8a9a0113","line":179,"in_reply_to":"7af24918_a460d436","updated":"2016-03-02 00:43:24.000000000","message":"Trying to auto-configure MTU is unrealistic.","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"0d5058eb4da468e9f49e11c3cf4a6021b514a213","unresolved":false,"context_lines":[{"line_number":176,"context_line":"               help\u003d_(\"This will choose the web framework in which to run \""},{"line_number":177,"context_line":"                      \"the Neutron API server. \u0027pecan\u0027 is a new experiemental \""},{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_228fad24","line":179,"in_reply_to":"7af24918_c28d297d","updated":"2016-03-01 17:12:03.000000000","message":"Neither I suggested you did. :)","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"e0e97a137ae8b82a1da87225e0bd697335b4ab4b","unresolved":false,"context_lines":[{"line_number":178,"context_line":"                      \"rewrite of the API server.\")),"},{"line_number":179,"context_line":"    cfg.IntOpt(\u0027global_physnet_mtu\u0027, default\u003d1500,"},{"line_number":180,"context_line":"               deprecated_name\u003d\u0027segment_mtu\u0027, deprecated_group\u003d\u0027ml2\u0027,"},{"line_number":181,"context_line":"               help\u003d_(\u0027MTU of the underlying physical network. Neutron uses\u0027"},{"line_number":182,"context_line":"                      \u0027this value to calculate MTU for all virtual network \u0027"},{"line_number":183,"context_line":"                      \u0027components. For flat and VLAN networks, neutron uses \u0027"},{"line_number":184,"context_line":"                      \u0027this value without modification. For overlay networks \u0027"}],"source_content_type":"text/x-python","patch_set":4,"id":"7af24918_a972887a","line":181,"updated":"2016-03-01 16:35:47.000000000","message":"missing space","commit_id":"45924d7f0047ddf49bde261355f255d3f53a865e"},{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"a263a9edcfb62f448a86eea956b4ae2cfef55f00","unresolved":false,"context_lines":[{"line_number":187,"context_line":"                      \u0027to 1500, the standard value for Ethernet.\u0027))"},{"line_number":188,"context_line":"]"},{"line_number":189,"context_line":""},{"line_number":190,"context_line":""},{"line_number":191,"context_line":"core_cli_opts \u003d ["},{"line_number":192,"context_line":"    cfg.StrOpt(\u0027state_path\u0027,"},{"line_number":193,"context_line":"               default\u003d\u0027/var/lib/neutron\u0027,"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_cc6ee0c8","line":190,"updated":"2016-03-08 04:30:30.000000000","message":"you don\u0027t need this space.","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"}],"neutron/plugins/common/utils.py":[{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"a263a9edcfb62f448a86eea956b4ae2cfef55f00","unresolved":false,"context_lines":[{"line_number":33,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":34,"context_line":""},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"def get_global_physnet_mtu():"},{"line_number":37,"context_line":"    \"\"\"Retrieves global MTU setting."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function so changing the config name won\u0027t"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_cca54005","line":36,"updated":"2016-03-08 04:30:30.000000000","message":"how many times do we have to fiddle with the mtu to get it right? I think this method is superfluous at this point, and I\u0027d vote to kill it.","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"dfdf6e28b1b4325e4df6797ad89a6ae62d2639ba","unresolved":false,"context_lines":[{"line_number":33,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":34,"context_line":""},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"def get_global_physnet_mtu():"},{"line_number":37,"context_line":"    \"\"\"Retrieves global MTU setting."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function so changing the config name won\u0027t"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_fcab69d4","line":36,"in_reply_to":"5aef4532_603b31d2","updated":"2016-03-08 18:34:59.000000000","message":"Don\u0027t get me wrong, the practice of encapsulation makes sense, if it was widely adopted for all config options, and in a more abstracted way.\n\nThis looks like a silly alias that matches the exact config name. What if we end up calling the config option universal_physnet_mtu? The method is gonna stink.","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"2952d16a147af2dcf2e6b3b51d13febc6dfc311c","unresolved":false,"context_lines":[{"line_number":33,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":34,"context_line":""},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"def get_global_physnet_mtu():"},{"line_number":37,"context_line":"    \"\"\"Retrieves global MTU setting."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function so changing the config name won\u0027t"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_df83ff02","line":36,"in_reply_to":"5aef4532_7c1c7977","updated":"2016-03-08 18:58:24.000000000","message":"Amen on that ^.","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"7c5e6a4648347ddc35d3332c91b0c4a8cdd1a025","unresolved":false,"context_lines":[{"line_number":33,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":34,"context_line":""},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"def get_global_physnet_mtu():"},{"line_number":37,"context_line":"    \"\"\"Retrieves global MTU setting."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function so changing the config name won\u0027t"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_603b31d2","line":36,"in_reply_to":"5aef4532_cca54005","updated":"2016-03-08 16:56:45.000000000","message":"Config options are not meant to be consumed as public interfaces. And this config option is now meant to be consumed by out of tree code. We *should* have the wrapper. [The position on that is also supported by oslo team.]","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"14e472b1de10e9e659550a0ef9b747e96bdec9c7","unresolved":false,"context_lines":[{"line_number":33,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":34,"context_line":""},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"def get_global_physnet_mtu():"},{"line_number":37,"context_line":"    \"\"\"Retrieves global MTU setting."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function so changing the config name won\u0027t"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_fcf40988","line":36,"in_reply_to":"5aef4532_fcab69d4","updated":"2016-03-08 18:42:20.000000000","message":"It may then *stink*, but it at least won\u0027t leave all out-of-tree plugins broken by the change just because you like tinkering your naming. I don\u0027t think it\u0027s fine to first tell someone to adopt an option as public API, then intentionally break them later. That said, I really prefer this patch goes in quick. If that\u0027s the culprit from it to happen here, then I lift my suggestion and suggest Kevin to respin as per your suggestion.","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"0d772a05a5202030c1e10c0e662d8f45c6e7596f","unresolved":false,"context_lines":[{"line_number":33,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":34,"context_line":""},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"def get_global_physnet_mtu():"},{"line_number":37,"context_line":"    \"\"\"Retrieves global MTU setting."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function so changing the config name won\u0027t"}],"source_content_type":"text/x-python","patch_set":6,"id":"5aef4532_7c1c7977","line":36,"in_reply_to":"5aef4532_fcf40988","updated":"2016-03-08 18:48:19.000000000","message":"Right, so let\u0027s make sure we choose a name we can reasonably be happy with, FOR EVER.","commit_id":"10ed6389f2a06f56a8792aa9e90c8656bf9b0fdd"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"3ad20a9a758b9d1d4272b0a4dd00454b0f5a790a","unresolved":false,"context_lines":[{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function to retrieve the MTU set by the operator"},{"line_number":40,"context_line":"    that is equal to or less than the MTU of their nodes\u0027 physical interfaces."},{"line_number":41,"context_line":"    Note that it is the responisbility of the plugin to deduct the value of"},{"line_number":42,"context_line":"    any encapsulation overhead required before advertising it to VMs."},{"line_number":43,"context_line":"    \"\"\""},{"line_number":44,"context_line":"    return cfg.CONF.global_physnet_mtu"}],"source_content_type":"text/x-python","patch_set":9,"id":"5aef4532_7fcbcb1e","line":41,"updated":"2016-03-08 19:05:01.000000000","message":"responsibility","commit_id":"3f376e884434bbc2b6e990052f5a76b9df80498e"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"e5f305b47dfc0bdf8674e95bfa7d4b028509899c","unresolved":false,"context_lines":[{"line_number":38,"context_line":""},{"line_number":39,"context_line":"    Plugins should use this function to retrieve the MTU set by the operator"},{"line_number":40,"context_line":"    that is equal to or less than the MTU of their nodes\u0027 physical interfaces."},{"line_number":41,"context_line":"    Note that it is the responisbility of the plugin to deduct the value of"},{"line_number":42,"context_line":"    any encapsulation overhead required before advertising it to VMs."},{"line_number":43,"context_line":"    \"\"\""},{"line_number":44,"context_line":"    return cfg.CONF.global_physnet_mtu"}],"source_content_type":"text/x-python","patch_set":9,"id":"5aef4532_8278fc90","line":41,"in_reply_to":"5aef4532_7fcbcb1e","updated":"2016-03-08 19:11:16.000000000","message":"Done","commit_id":"3f376e884434bbc2b6e990052f5a76b9df80498e"}],"neutron/tests/unit/common/test_config.py":[{"author":{"_account_id":8124,"name":"cbrandily","email":"zzelle@gmail.com","username":"cbrandily"},"change_message_id":"24086011895e6192c9433d0208dcf3adf19c6969","unresolved":false,"context_lines":[{"line_number":26,"context_line":"        cfg.CONF.set_override(\u0027segment_mtu\u0027, 1600, group\u003d\u0027ml2\u0027)"},{"line_number":27,"context_line":"        self.assertEqual(1600, cfg.CONF.global_physnet_mtu)"},{"line_number":28,"context_line":"        # if specified, it no longer inherits"},{"line_number":29,"context_line":"        cfg.CONF.set_override(\u0027global_physnet_mtu\u0027, 1700)"},{"line_number":30,"context_line":"        cfg.CONF.set_override(\u0027segment_mtu\u0027, 1800, group\u003d\u0027ml2\u0027)"},{"line_number":31,"context_line":"        self.assertEqual(1700, cfg.CONF.global_physnet_mtu)"}],"source_content_type":"text/x-python","patch_set":1,"id":"9aed3d3a_ed114cb1","line":29,"range":{"start_line":29,"start_character":17,"end_line":29,"end_character":29},"updated":"2016-02-25 21:15:42.000000000","message":"it would simplify tests","commit_id":"9353aadb0e582a59b051f2f9a650683aa7994fc2"}],"releasenotes/notes/segment_mtu_to_global_physnet_mtu-9cee5ff09557edeb.yaml":[{"author":{"_account_id":9515,"name":"Matthew Kassawara","email":"mkassawara@gmail.com","username":"ionosphere80"},"change_message_id":"294aa7191997a4f2bf8df56c9d302c82fdbea855","unresolved":false,"context_lines":[{"line_number":4,"context_line":"    deprecated and replaced with the \u0027global_physnet_mtu\u0027 option"},{"line_number":5,"context_line":"    in the main Neutron configuration. This option is meant to"},{"line_number":6,"context_line":"    be used by all plugins for an operator to set their physical"},{"line_number":7,"context_line":"    network\u0027s MTU, regardless of the backend plugin."}],"source_content_type":"text/x-yaml","patch_set":8,"id":"5aef4532_3c9cb168","line":7,"updated":"2016-03-08 18:39:32.000000000","message":"Suggestion: Use \"reference\" instead of \"set\" to eliminate the potential of people thinking neutron sets MTUs for physical network components. Should probably also reference the virtual network components as I did in the commit message.","commit_id":"a75439ef8d03e10800c6a154b5d6a8f9f5f5a21b"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"3b6dd259c925ee9817b0e6da9abf6760923ae3c1","unresolved":false,"context_lines":[{"line_number":4,"context_line":"    deprecated and replaced with the \u0027global_physnet_mtu\u0027 option"},{"line_number":5,"context_line":"    in the main Neutron configuration. This option is meant to"},{"line_number":6,"context_line":"    be used by all plugins for an operator to set their physical"},{"line_number":7,"context_line":"    network\u0027s MTU, regardless of the backend plugin."}],"source_content_type":"text/x-yaml","patch_set":8,"id":"5aef4532_5fa1afa2","line":7,"in_reply_to":"5aef4532_3c9cb168","updated":"2016-03-08 18:58:02.000000000","message":"Done","commit_id":"a75439ef8d03e10800c6a154b5d6a8f9f5f5a21b"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"3ad20a9a758b9d1d4272b0a4dd00454b0f5a790a","unresolved":false,"context_lines":[{"line_number":4,"context_line":"    deprecated and replaced with the \u0027global_physnet_mtu\u0027 option"},{"line_number":5,"context_line":"    in the main Neutron configuration. This option is meant to"},{"line_number":6,"context_line":"    be used by all plugins for an operator to reference their"},{"line_number":7,"context_line":"    physical network\u0027s MTU, regardless of the backend plugin."}],"source_content_type":"text/x-yaml","patch_set":9,"id":"5aef4532_dfa01f46","line":7,"updated":"2016-03-08 19:05:01.000000000","message":"better have some wording that would not suggest direct usage of the option in plugins.","commit_id":"3f376e884434bbc2b6e990052f5a76b9df80498e"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"e5f305b47dfc0bdf8674e95bfa7d4b028509899c","unresolved":false,"context_lines":[{"line_number":4,"context_line":"    deprecated and replaced with the \u0027global_physnet_mtu\u0027 option"},{"line_number":5,"context_line":"    in the main Neutron configuration. This option is meant to"},{"line_number":6,"context_line":"    be used by all plugins for an operator to reference their"},{"line_number":7,"context_line":"    physical network\u0027s MTU, regardless of the backend plugin."}],"source_content_type":"text/x-yaml","patch_set":9,"id":"5aef4532_e2f460f3","line":7,"in_reply_to":"5aef4532_dfa01f46","updated":"2016-03-08 19:11:16.000000000","message":"Done","commit_id":"3f376e884434bbc2b6e990052f5a76b9df80498e"}]}