)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"68cd69f4cee261c4f68fb7eccd42a80f0ea2940e","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Revert \"Deprecate \u0027ovs_use_veth\u0027 and \u0027veth_mtu\u0027 options\""},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The reason for reverting this is that when we tested this OVS port is show as down and it can actually send traffic. An explanation is below:"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"The DHCP port has \"tap\" in the name, but doesn\u0027t appear to be a tap device.  It appears to be an OVS internal device.  That device was then put into a different network namespace, which isn\u0027t visible to the root namespace in which ovs-vswitchd is running.  Using internal devices in this manor is kind of a hack since ovs-vswitchd can\u0027t actually see the device, but it can still send and receive traffic on it."},{"line_number":12,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"1a122d0e_08cba054","line":9,"updated":"2016-04-18 06:08:13.000000000","message":"Where is it showing as down? When I use a tap device locally in dev it seems to show as ACTIVE.","commit_id":"3b10f06ef859fcc2f137c310cf68fe18d58d8547"},{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"95c7d037ec2f0fb883333aca22d3d6cb8a1bcd6d","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Revert \"Deprecate \u0027ovs_use_veth\u0027 and \u0027veth_mtu\u0027 options\""},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The reason for reverting this is that when we tested this OVS port is show as down and it can actually send traffic. An explanation is below:"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"The DHCP port has \"tap\" in the name, but doesn\u0027t appear to be a tap device.  It appears to be an OVS internal device.  That device was then put into a different network namespace, which isn\u0027t visible to the root namespace in which ovs-vswitchd is running.  Using internal devices in this manor is kind of a hack since ovs-vswitchd can\u0027t actually see the device, but it can still send and receive traffic on it."},{"line_number":12,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"1a122d0e_7b55b74e","line":9,"in_reply_to":"1a122d0e_08cba054","updated":"2016-04-18 08:59:22.000000000","message":"the tap device is up but on the ovs the link state is down","commit_id":"3b10f06ef859fcc2f137c310cf68fe18d58d8547"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"68cd69f4cee261c4f68fb7eccd42a80f0ea2940e","unresolved":false,"context_lines":[{"line_number":7,"context_line":"Revert \"Deprecate \u0027ovs_use_veth\u0027 and \u0027veth_mtu\u0027 options\""},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The reason for reverting this is that when we tested this OVS port is show as down and it can actually send traffic. An explanation is below:"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"The DHCP port has \"tap\" in the name, but doesn\u0027t appear to be a tap device.  It appears to be an OVS internal device.  That device was then put into a different network namespace, which isn\u0027t visible to the root namespace in which ovs-vswitchd is running.  Using internal devices in this manor is kind of a hack since ovs-vswitchd can\u0027t actually see the device, but it can still send and receive traffic on it."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This reverts commit 850b4c025fa5652bfa474c78e7775d74fdb93df6."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"1a122d0e_a8870c0e","line":10,"updated":"2016-04-18 06:08:13.000000000","message":"can you wordwrap this commit message a bit?","commit_id":"3b10f06ef859fcc2f137c310cf68fe18d58d8547"},{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"95c7d037ec2f0fb883333aca22d3d6cb8a1bcd6d","unresolved":false,"context_lines":[{"line_number":7,"context_line":"Revert \"Deprecate \u0027ovs_use_veth\u0027 and \u0027veth_mtu\u0027 options\""},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The reason for reverting this is that when we tested this OVS port is show as down and it can actually send traffic. An explanation is below:"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"The DHCP port has \"tap\" in the name, but doesn\u0027t appear to be a tap device.  It appears to be an OVS internal device.  That device was then put into a different network namespace, which isn\u0027t visible to the root namespace in which ovs-vswitchd is running.  Using internal devices in this manor is kind of a hack since ovs-vswitchd can\u0027t actually see the device, but it can still send and receive traffic on it."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This reverts commit 850b4c025fa5652bfa474c78e7775d74fdb93df6."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"1a122d0e_5b50bb5e","line":10,"in_reply_to":"1a122d0e_a8870c0e","updated":"2016-04-18 08:59:22.000000000","message":"Done","commit_id":"3b10f06ef859fcc2f137c310cf68fe18d58d8547"},{"author":{"_account_id":7787,"name":"Kevin Benton","email":"kevin@benton.pub","username":"blak111"},"change_message_id":"68cd69f4cee261c4f68fb7eccd42a80f0ea2940e","unresolved":false,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The reason for reverting this is that when we tested this OVS port is show as down and it can actually send traffic. An explanation is below:"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"The DHCP port has \"tap\" in the name, but doesn\u0027t appear to be a tap device.  It appears to be an OVS internal device.  That device was then put into a different network namespace, which isn\u0027t visible to the root namespace in which ovs-vswitchd is running.  Using internal devices in this manor is kind of a hack since ovs-vswitchd can\u0027t actually see the device, but it can still send and receive traffic on it."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This reverts commit 850b4c025fa5652bfa474c78e7775d74fdb93df6."},{"line_number":14,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"1a122d0e_48ae3886","line":11,"updated":"2016-04-18 06:08:13.000000000","message":"it is a tap device, it just gets moved into the namespace as you observed. Is there a bug that prevents this from working correctly?","commit_id":"3b10f06ef859fcc2f137c310cf68fe18d58d8547"},{"author":{"_account_id":841,"name":"Akihiro Motoki","email":"amotoki@gmail.com","username":"amotoki"},"change_message_id":"46ac7a105ad020524658a7296d1b63b8f39ceeee","unresolved":false,"context_lines":[{"line_number":7,"context_line":"Revert \"Deprecate \u0027ovs_use_veth\u0027 and \u0027veth_mtu\u0027 options\""},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The reason for reverting this is that when we tested this OVS"},{"line_number":10,"context_line":"port is show as DOWN and it can actually send traffic."},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"An explanation is below:"},{"line_number":13,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"1a122d0e_fdbff25d","line":10,"updated":"2016-04-18 12:07:40.000000000","message":"I think ovs_use_veth is still usrful and agree this revert.\n\nIt also prevents a network controller from determining whether the port is actually enabled based on reported port status.\n\nI also saw a problem with older OVS (shipped with Ubuntu 14.04 LTS) with l3-agent that FIP on OVS internal ports sometimes did not respond to ICMP request. ovs_use_veth\u003dTrue solves the problem.","commit_id":"dab6070cafb95308cb8580856ecfdf213e8f77b8"},{"author":{"_account_id":841,"name":"Akihiro Motoki","email":"amotoki@gmail.com","username":"amotoki"},"change_message_id":"46ac7a105ad020524658a7296d1b63b8f39ceeee","unresolved":false,"context_lines":[{"line_number":15,"context_line":"a tap device. It appears to be an OVS internal device. That device"},{"line_number":16,"context_line":"was then put into a different network namespace, which isn\u0027t"},{"line_number":17,"context_line":"visible to the root namespace in which ovs-vswitchd is running."},{"line_number":18,"context_line":"Using internal devices in this manor is kind of a hack since"},{"line_number":19,"context_line":"ovs-vswitchd can\u0027t actually see the device, but it can still"},{"line_number":20,"context_line":"send and receive traffic on it."},{"line_number":21,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"1a122d0e_98ab6cd6","line":18,"range":{"start_line":18,"start_character":31,"end_line":18,"end_character":37},"updated":"2016-04-18 12:07:40.000000000","message":"manner","commit_id":"dab6070cafb95308cb8580856ecfdf213e8f77b8"}]}
