)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":7,"name":"Jay Pipes","email":"jaypipes@gmail.com","username":"jaypipes"},"change_message_id":"5e055bfc0e767ffb7863664d3541cbeed56200ed","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"vif_plug_ovs: Always set MTU when plugging devices"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"neutron allows for customizable MTUs, while nova allows for in-place"},{"line_number":10,"context_line":"upgrades. Recent versions of neutron have made significant changes to"},{"line_number":11,"context_line":"how MTUs are calculated and a side effect of this is that possibility"},{"line_number":12,"context_line":"that the network MTU can change. While we don\u0027t recommend doing such a"},{"line_number":13,"context_line":"thing, it does happen we can make life a little easier for the users who"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_df8fd572","line":10,"range":{"start_line":9,"start_character":0,"end_line":10,"end_character":9},"updated":"2017-01-06 17:39:07.000000000","message":"This first sentence doesn\u0027t make sense to me. Or rather, I don\u0027t see the connection between Neutron allowing customizable MTUs and Nova allowing for in-place upgrades...","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"b060b88e0259147dede2e3dd330433c6cf50fbf8","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"vif_plug_ovs: Always set MTU when plugging devices"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"neutron allows for customizable MTUs, while nova allows for in-place"},{"line_number":10,"context_line":"upgrades. Recent versions of neutron have made significant changes to"},{"line_number":11,"context_line":"how MTUs are calculated and a side effect of this is that possibility"},{"line_number":12,"context_line":"that the network MTU can change. While we don\u0027t recommend doing such a"},{"line_number":13,"context_line":"thing, it does happen we can make life a little easier for the users who"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_512e79fb","line":10,"range":{"start_line":9,"start_character":0,"end_line":10,"end_character":9},"in_reply_to":"ba5201f7_df8fd572","updated":"2017-01-06 18:05:10.000000000","message":"Yeah, I know what I mean but it\u0027s not clear. Let me take another shot at it.","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"c3ea9385040007411c25bc383384486be8bc026a","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"vif_plug_ovs: Always set MTU when plugging devices"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"neutron allows for customizable MTUs, while nova allows for in-place"},{"line_number":10,"context_line":"upgrades. Recent versions of neutron have made significant changes to"},{"line_number":11,"context_line":"how MTUs are calculated and a side effect of this is that possibility"},{"line_number":12,"context_line":"that the network MTU can change. While we don\u0027t recommend doing such a"},{"line_number":13,"context_line":"thing, it does happen we can make life a little easier for the users who"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_51e039cd","line":10,"range":{"start_line":9,"start_character":0,"end_line":10,"end_character":9},"in_reply_to":"ba5201f7_df8fd572","updated":"2017-01-06 18:10:25.000000000","message":"in mitaka when the mtu was change form 0 to 1500 by default it caused issues when an in-place upgrade was done without evacuating the host as the veth pairs would not be updated.\n\nthis also happens if a custom mtu was configured for neutron to use. \n\ni belive stephen encountered this issue because some deployment tools set a custom mtu of 1400 and which was recently changed and on upgrade they also had the issue that mtus of the veth pairs were not updated.","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":7,"name":"Jay Pipes","email":"jaypipes@gmail.com","username":"jaypipes"},"change_message_id":"5e055bfc0e767ffb7863664d3541cbeed56200ed","unresolved":false,"context_lines":[{"line_number":9,"context_line":"neutron allows for customizable MTUs, while nova allows for in-place"},{"line_number":10,"context_line":"upgrades. Recent versions of neutron have made significant changes to"},{"line_number":11,"context_line":"how MTUs are calculated and a side effect of this is that possibility"},{"line_number":12,"context_line":"that the network MTU can change. While we don\u0027t recommend doing such a"},{"line_number":13,"context_line":"thing, it does happen we can make life a little easier for the users who"},{"line_number":14,"context_line":"do by configuring as much of the network \"plumbing\" to use the new MTU"},{"line_number":15,"context_line":"when plugging interfaces."},{"line_number":16,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_ffb81913","line":13,"range":{"start_line":12,"start_character":33,"end_line":13,"end_character":5},"updated":"2017-01-06 17:39:07.000000000","message":"What don\u0027t we recommend doing exactly? Changing the MTU of the network segment? If we don\u0027t recommend this, why do we allow it?","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"b060b88e0259147dede2e3dd330433c6cf50fbf8","unresolved":false,"context_lines":[{"line_number":9,"context_line":"neutron allows for customizable MTUs, while nova allows for in-place"},{"line_number":10,"context_line":"upgrades. Recent versions of neutron have made significant changes to"},{"line_number":11,"context_line":"how MTUs are calculated and a side effect of this is that possibility"},{"line_number":12,"context_line":"that the network MTU can change. While we don\u0027t recommend doing such a"},{"line_number":13,"context_line":"thing, it does happen we can make life a little easier for the users who"},{"line_number":14,"context_line":"do by configuring as much of the network \"plumbing\" to use the new MTU"},{"line_number":15,"context_line":"when plugging interfaces."},{"line_number":16,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_b1008567","line":13,"range":{"start_line":12,"start_character":33,"end_line":13,"end_character":5},"in_reply_to":"ba5201f7_ffb81913","updated":"2017-01-06 18:05:10.000000000","message":"Because the world isn\u0027t perfect :) You shouldn\u0027t be changing the network MTU willy nilly but it does happen, for example during live upgrades (neutron has done a lot of work around MTUs lately). I\u0027ll clarify this.","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"c3ea9385040007411c25bc383384486be8bc026a","unresolved":false,"context_lines":[{"line_number":9,"context_line":"neutron allows for customizable MTUs, while nova allows for in-place"},{"line_number":10,"context_line":"upgrades. Recent versions of neutron have made significant changes to"},{"line_number":11,"context_line":"how MTUs are calculated and a side effect of this is that possibility"},{"line_number":12,"context_line":"that the network MTU can change. While we don\u0027t recommend doing such a"},{"line_number":13,"context_line":"thing, it does happen we can make life a little easier for the users who"},{"line_number":14,"context_line":"do by configuring as much of the network \"plumbing\" to use the new MTU"},{"line_number":15,"context_line":"when plugging interfaces."},{"line_number":16,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_d14269ad","line":13,"range":{"start_line":12,"start_character":33,"end_line":13,"end_character":5},"in_reply_to":"ba5201f7_ffb81913","updated":"2017-01-06 18:10:25.000000000","message":"doing an in place upgrade.\nthe recommendation would be to evacuate the compute host before upgrading it to the new openstack release.","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":7,"name":"Jay Pipes","email":"jaypipes@gmail.com","username":"jaypipes"},"change_message_id":"5e055bfc0e767ffb7863664d3541cbeed56200ed","unresolved":false,"context_lines":[{"line_number":16,"context_line":""},{"line_number":17,"context_line":"This still leaves the user with the task of setting the MTU inside the"},{"line_number":18,"context_line":"guest themselves, or waiting for the DHCP lease to expire so that the"},{"line_number":19,"context_line":"new MTU will be pulled in."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Change-Id: If09eda334cddc74910dda7a4fb498b7987714be3"},{"line_number":22,"context_line":"Closes-bug: #1649845"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_5f5765c9","line":19,"updated":"2017-01-06 17:39:07.000000000","message":"If this is the case, I\u0027m not entirely sure what the value of doing this in os-vif is? I mean, if the user still needs to perform manual steps in the guest...","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"b060b88e0259147dede2e3dd330433c6cf50fbf8","unresolved":false,"context_lines":[{"line_number":16,"context_line":""},{"line_number":17,"context_line":"This still leaves the user with the task of setting the MTU inside the"},{"line_number":18,"context_line":"guest themselves, or waiting for the DHCP lease to expire so that the"},{"line_number":19,"context_line":"new MTU will be pulled in."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Change-Id: If09eda334cddc74910dda7a4fb498b7987714be3"},{"line_number":22,"context_line":"Closes-bug: #1649845"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_11c4f13d","line":19,"in_reply_to":"ba5201f7_5f5765c9","updated":"2017-01-06 18:05:10.000000000","message":"I should reword this as \"The guest will continue to see dropped packets until...\". If the user is happy with this, they can wait for DHCP to intervene (which is what I imagine most folks will do)","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"c3ea9385040007411c25bc383384486be8bc026a","unresolved":false,"context_lines":[{"line_number":16,"context_line":""},{"line_number":17,"context_line":"This still leaves the user with the task of setting the MTU inside the"},{"line_number":18,"context_line":"guest themselves, or waiting for the DHCP lease to expire so that the"},{"line_number":19,"context_line":"new MTU will be pulled in."},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Change-Id: If09eda334cddc74910dda7a4fb498b7987714be3"},{"line_number":22,"context_line":"Closes-bug: #1649845"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"ba5201f7_71cd3d30","line":19,"in_reply_to":"ba5201f7_5f5765c9","updated":"2017-01-06 18:10:25.000000000","message":"the dhcp lease refresh will updated the guest at half the remaining interval automatically.\nif the operator has increassed the mtu e.g. the enabled jumbo frames then the guest would continue to function with the smaller old mtu untill updated by dhcp.\n\nif the operator reduced the mtu it would cause network issues for the guests. in this case it is operator error in my opipion as that is not a safe opperation as it might break the tenant workload.","commit_id":"233ac49d4a65984111fab960dfba3a60984646e4"},{"author":{"_account_id":7,"name":"Jay Pipes","email":"jaypipes@gmail.com","username":"jaypipes"},"change_message_id":"5aed0f70bd7e206a11a74b99e36ee1510a8b7ddc","unresolved":false,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Recent versions of neutron have made significant changes to how MTUs"},{"line_number":10,"context_line":"are calculated. This work appears to have been completed and it\u0027s had"},{"line_number":11,"context_line":"knock-on effects for some installers [1] which have changed their"},{"line_number":12,"context_line":"default MTUs accordingly. In general, it is not recommended that one"},{"line_number":13,"context_line":"change the MTU of a network segment in-flight as it can result in"},{"line_number":14,"context_line":"dropped packets (depending on the change). However, the combination of"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":7,"id":"ba5201f7_94e60fe6","line":11,"updated":"2017-01-06 18:34:54.000000000","message":"did you forget to include a footnote here? :)","commit_id":"4b741299a31018c55a13884030e3736cb341d07e"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"4045c2b88a63ddac02eb5a648901e3669c565283","unresolved":false,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Recent versions of neutron have made significant changes to how MTUs"},{"line_number":10,"context_line":"are calculated. This work appears to have been completed and it\u0027s had"},{"line_number":11,"context_line":"knock-on effects for some installers [1] which have changed their"},{"line_number":12,"context_line":"default MTUs accordingly. In general, it is not recommended that one"},{"line_number":13,"context_line":"change the MTU of a network segment in-flight as it can result in"},{"line_number":14,"context_line":"dropped packets (depending on the change). However, the combination of"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":7,"id":"ba5201f7_b32ea2f8","line":11,"in_reply_to":"ba5201f7_94e60fe6","updated":"2017-01-09 09:48:46.000000000","message":"Done and done :)","commit_id":"4b741299a31018c55a13884030e3736cb341d07e"}],"vif_plug_ovs/linux_net.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"a34f1918475ce1e51abb5891e7f91c99829ebde5","unresolved":false,"context_lines":[{"line_number":168,"context_line":"    processutils.execute(\u0027brctl\u0027, \u0027addif\u0027, bridge, dev)"},{"line_number":169,"context_line":""},{"line_number":170,"context_line":""},{"line_number":171,"context_line":"def _update_device_mtu(dev, mtu, interface_type\u003dNone):"},{"line_number":172,"context_line":"    # Note at present there is no support for setting the"},{"line_number":173,"context_line":"    # mtu for vhost-user type ports."},{"line_number":174,"context_line":"    if mtu and interface_type not in ["},{"line_number":175,"context_line":"            constants.OVS_VHOSTUSER_INTERFACE_TYPE,"},{"line_number":176,"context_line":"            constants.OVS_VHOSTUSER_CLIENT_INTERFACE_TYPE]:"},{"line_number":177,"context_line":"        _set_device_mtu(dev, mtu)"},{"line_number":178,"context_line":"    else:"},{"line_number":179,"context_line":"        LOG.debug(\"MTU not set on %(interface_name)s interface \""},{"line_number":180,"context_line":"                  \"of type %(interface_type)s.\","},{"line_number":181,"context_line":"                  {\u0027interface_name\u0027: dev,"},{"line_number":182,"context_line":"                   \u0027interface_type\u0027: interface_type})"},{"line_number":183,"context_line":""},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"def _set_device_mtu(dev, mtu):"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a6eadb0_5b3d92c2","line":182,"range":{"start_line":171,"start_character":0,"end_line":182,"end_character":53},"updated":"2016-12-16 16:45:21.000000000","message":"this will conflict with my vhost-user mtu change so ill rebase my change on top of this patch.","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"82a7d8d06b4058bdeac81f0ba513d13332415d26","unresolved":false,"context_lines":[{"line_number":184,"context_line":""},{"line_number":185,"context_line":"def _set_device_mtu(dev, mtu):"},{"line_number":186,"context_line":"    \"\"\"Set the device MTU.\"\"\""},{"line_number":187,"context_line":"    processutils.execute(\u0027ip\u0027, \u0027link\u0027, \u0027set\u0027, dev, \u0027mtu\u0027, mtu,"},{"line_number":188,"context_line":"                         check_exit_code\u003d[0, 2, 254])"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a6eadb0_9d4aba53","line":188,"range":{"start_line":187,"start_character":5,"end_line":188,"end_character":53},"updated":"2016-12-16 19:32:06.000000000","message":"the equivalent command on windows would be \n\u0027netsh\u0027, \u0027interface\u0027, \u0027ipv4\u0027, \u0027set\u0027, \u0027subinterface\u0027, dev, \u0027mtu\u003d%s\u0027 % mtu, \u0027store\u003dpersistent\u0027\n\nsee http://www.sysadmintutorials.com/tutorials/microsoft/windows-2008-r2/how-to-set-windows-2008-r2-mtu/\n\nagain i dont know what effect this has when used with the windows kernel ovs datapath.\n\ni would suggest moving windows supprot to a seperate patch\nas we might need to do it via ovs instead like vhost-user or both","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"}],"vif_plug_ovs/ovs.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3ffa073a3bb2c868407490b4af98e74c67eca357","unresolved":false,"context_lines":[{"line_number":146,"context_line":"    def _plug_vif_windows(self, vif, instance_info):"},{"line_number":147,"context_line":"        \"\"\"Create a per-VIF OVS port.\"\"\""},{"line_number":148,"context_line":""},{"line_number":149,"context_line":"        if not linux_net.device_exists(vif.id):"},{"line_number":150,"context_line":"            linux_net.ensure_ovs_bridge(vif.network.bridge,"},{"line_number":151,"context_line":"                                            constants.OVS_DATAPATH_SYSTEM)"},{"line_number":152,"context_line":"            self._create_vif_port(vif, vif.id, instance_info)"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a6eadb0_3616eda7","line":149,"range":{"start_line":149,"start_character":8,"end_line":149,"end_character":47},"updated":"2016-12-19 15:33:21.000000000","message":"this is horribly broken...\nit is allway true as device exists\nis just \nos.path.exists(\u0027/sys/class/net/%s\u0027 % device)\n\nthis only is false if c:\\sys\\calss\\net\\\u003cdevice\u003e exists which is very unlikely and also probably invalid logic even if found.\n\nthis only works because ensure_ovs_bridge and _create_vif_port can always be called safely regardless of if the port or bridge exists.\n\nas a result, your else block will never be executed.\n\nwe need to actually refactor this so that we have a set of libs for networking.\none for ovs\nand one with two derived implementations for ip commands(linux and windows)\n\nill do that before ocata comes out if i have time.","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"4c3f4a291bd589d22b5abcfea18a07d33830ba05","unresolved":false,"context_lines":[{"line_number":150,"context_line":"            linux_net.ensure_ovs_bridge(vif.network.bridge,"},{"line_number":151,"context_line":"                                            constants.OVS_DATAPATH_SYSTEM)"},{"line_number":152,"context_line":"            self._create_vif_port(vif, vif.id, instance_info)"},{"line_number":153,"context_line":"        else:"},{"line_number":154,"context_line":"            self._update_vif_port(vif, vif.id)"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"    def plug(self, vif, instance_info):"},{"line_number":157,"context_line":"        if not hasattr(vif, \"port_profile\"):"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a6eadb0_029bab18","line":154,"range":{"start_line":153,"start_character":8,"end_line":154,"end_character":45},"updated":"2016-12-16 19:19:58.000000000","message":"actully this is wronge\nwe cannot call this on windows as it dose not have ip tool to update the vif mtu. we would have to extend linux_net._set_device_mtu to use netsh on window to set the mtu. even then im not sure if that works or not so this line should be deleted i think.","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"94352ca121c024da15426e9c857d53fc6e451124","unresolved":false,"context_lines":[{"line_number":150,"context_line":"            linux_net.ensure_ovs_bridge(vif.network.bridge,"},{"line_number":151,"context_line":"                                            constants.OVS_DATAPATH_SYSTEM)"},{"line_number":152,"context_line":"            self._create_vif_port(vif, vif.id, instance_info)"},{"line_number":153,"context_line":"        else:"},{"line_number":154,"context_line":"            self._update_vif_port(vif, vif.id)"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"    def plug(self, vif, instance_info):"},{"line_number":157,"context_line":"        if not hasattr(vif, \"port_profile\"):"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a6eadb0_6480c147","line":154,"range":{"start_line":153,"start_character":8,"end_line":154,"end_character":45},"in_reply_to":"1a6eadb0_029bab18","updated":"2016-12-19 09:16:39.000000000","message":"But we\u0027re not using Windows kernel networking - we\u0027re using OVS? I added this because I noted that the create path sets the MTU (_create_vif_port -\u003e linux_net.create_ovs_vif_port -\u003e linux_net._update_device_mtu)","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"3ffa073a3bb2c868407490b4af98e74c67eca357","unresolved":false,"context_lines":[{"line_number":150,"context_line":"            linux_net.ensure_ovs_bridge(vif.network.bridge,"},{"line_number":151,"context_line":"                                            constants.OVS_DATAPATH_SYSTEM)"},{"line_number":152,"context_line":"            self._create_vif_port(vif, vif.id, instance_info)"},{"line_number":153,"context_line":"        else:"},{"line_number":154,"context_line":"            self._update_vif_port(vif, vif.id)"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"    def plug(self, vif, instance_info):"},{"line_number":157,"context_line":"        if not hasattr(vif, \"port_profile\"):"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a6eadb0_a30ba12a","line":154,"range":{"start_line":153,"start_character":8,"end_line":154,"end_character":45},"in_reply_to":"1a6eadb0_6480c147","updated":"2016-12-19 15:33:21.000000000","message":"there is a windows kernel ovs datapath implementation that is used with HyperV.\nya like above that is a bug in the current code.\n\nwhen we call _update_device_mtu it calls\n_set_device_mtu which does \n\nprocessutils.execute(\u0027ip\u0027, \u0027link\u0027, \u0027set\u0027, dev, \u0027mtu\u0027, mtu,\n                         check_exit_code\u003d[0, 2, 254])\n\nif you install oslo_concurrency on windows and test what happens it exits with exit code 2, file not found.\n\nfrom oslo_concurrency import processutils\n\u003e\u003e\u003e processutils.execute(\u0027ip\u0027,check_exit_code\u003d[0, 2, 254])\nTraceback (most recent call last):\n  File \"\u003cstdin\u003e\", line 1, in \u003cmodule\u003e\n  File \"/usr/lib/python2.7/site-packages/oslo.concurrency-3.16.0-py2.7.egg/oslo_concurrency/processutils.py\", line 368, in execute\n    env\u003denv_variables)\n  File \"/usr/lib/python2.7/subprocess.py\", line 711, in __init__\n    errread, errwrite)\n  File \"/usr/lib/python2.7/subprocess.py\", line 1343, in _execute_child\n    raise child_exception\nOSError: [Errno 2] No such file or directory\n\nbecause errno 2 is allowed in the check_exit_code it does not fail and pluggin works.\n\nwe need to get a real gate in place to do more then unit testing. ill talk to waldek about triggering the intel ci on changes for pike but we should have some dsvm gate jobs also.\n\nso technically this line will not break anything because 1 it will never be called and 2 if it was even though it generates an error and does notting that is one of the allowed errors so nothing breaks.","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"8663c9a60bb3e611641eef3ea85bfc9b9e539910","unresolved":false,"context_lines":[{"line_number":150,"context_line":"            linux_net.ensure_ovs_bridge(vif.network.bridge,"},{"line_number":151,"context_line":"                                            constants.OVS_DATAPATH_SYSTEM)"},{"line_number":152,"context_line":"            self._create_vif_port(vif, vif.id, instance_info)"},{"line_number":153,"context_line":"        else:"},{"line_number":154,"context_line":"            self._update_vif_port(vif, vif.id)"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"    def plug(self, vif, instance_info):"},{"line_number":157,"context_line":"        if not hasattr(vif, \"port_profile\"):"}],"source_content_type":"text/x-python","patch_set":5,"id":"ba5201f7_1c7fe166","line":154,"range":{"start_line":153,"start_character":8,"end_line":154,"end_character":45},"in_reply_to":"1a6eadb0_a30ba12a","updated":"2017-01-05 12:36:39.000000000","message":"I\u0027ve removed all of this. Windows support can come later","commit_id":"a4e505972531fc095fadfba88d5512ddbb7ac5b0"}],"vif_plug_ovs/tests/test_plugin.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7134ade1e336e368c1ad02bd91a1a5d74e5df67f","unresolved":false,"context_lines":[{"line_number":215,"context_line":"        self.assertFalse(_update_vif_port.called)"},{"line_number":216,"context_line":"        ensure_ovs_bridge.assert_has_calls(calls[\u0027ensure_ovs_bridge\u0027])"},{"line_number":217,"context_line":""},{"line_number":218,"context_line":"        # reset call stacks"},{"line_number":219,"context_line":""},{"line_number":220,"context_line":"        _create_vif_port.reset_mock()"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"        # plugging existing devices should result in devices being updated"},{"line_number":223,"context_line":""},{"line_number":224,"context_line":"        device_exists.return_value \u003d True"},{"line_number":225,"context_line":"        plugin.plug(vif, self.instance)"},{"line_number":226,"context_line":"        self.assertFalse(_create_vif_port.called)"},{"line_number":227,"context_line":"        _update_vif_port.assert_has_calls(calls[\u0027_update_vif_port\u0027])"},{"line_number":228,"context_line":""},{"line_number":229,"context_line":"    def test_plug_ovs_windows(self):"},{"line_number":230,"context_line":"        self._check_plug_ovs_windows(self.vif_ovs)"}],"source_content_type":"text/x-python","patch_set":3,"id":"1a6eadb0_6ebb4073","line":227,"range":{"start_line":218,"start_character":7,"end_line":227,"end_character":68},"updated":"2016-12-13 01:51:35.000000000","message":"this looks good to me","commit_id":"1c4a9f8361e9b54beec7c24f78bf54e5d1e8a40f"}]}
