)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"9345e7ad22ecb9e3495717d5257952b774480303","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"02ddc48d_8fa62988","updated":"2021-12-14 14:04:45.000000000","message":"I believe it would be really good to add a regression avoidance test here, but I don\u0027t see yet how.\n\nManually I\u0027m testing after adding a sleep() like this:\nhttps://review.opendev.org/c/openstack/neutron/+/820896\n\nBut that\u0027s clearly not something we can do in an automated test. Please let me know if you see a way.","commit_id":"69e311cb749892f63d1050d8c8921cf61db6eea2"},{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"1882c6948876a161c122f2c4978b221facadaf5d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":9,"id":"a0cbc19b_72ebbdc6","updated":"2021-12-21 12:13:40.000000000","message":"recheck","commit_id":"ac0381f5261a963a0a4e7a6b9e693e2150965417"},{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"5b9fd2d926f3043b9ae8f5c07d21eced6a81fea6","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":10,"id":"1d90c3b6_829a8a23","updated":"2021-12-22 15:18:30.000000000","message":"Hopefully fixed many of the failing functional tests.","commit_id":"c788962f31eb1d750ec164c0175ab1e3e1103f71"},{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"553f1aea66538ebf73390f7a8c26323da98859bf","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":11,"id":"5e9c6e43_863b0e5a","updated":"2022-01-03 12:44:10.000000000","message":"Hopefully fixed most fullstack test failures (maybe all but test_ovs_dhcp_agent_extension_ping_vms).","commit_id":"97e1c85aaa6038c01b532f86c13454d82f57a55d"},{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"2206570466f4d09671696e2f9eb79fe5d33c526e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":12,"id":"c1c0cbd8_411d5e7c","updated":"2022-01-07 12:04:50.000000000","message":"Good catch!\n\nPlease take care of the comment in interface.py to avoid repeating the issue we had dome releases ago.","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"}],"neutron/agent/linux/interface.py":[{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"2206570466f4d09671696e2f9eb79fe5d33c526e","unresolved":true,"context_lines":[{"line_number":258,"context_line":"        ip_lib.sysctl(cmd, namespace\u003dnamespace)"},{"line_number":259,"context_line":""},{"line_number":260,"context_line":"    @abc.abstractmethod"},{"line_number":261,"context_line":"    def plug_new(self, network_id, port_id, device_name, mac_address,"},{"line_number":262,"context_line":"                 bridge\u003dNone, namespace\u003dNone, prefix\u003dNone, mtu\u003dNone,"},{"line_number":263,"context_line":"                 link_up\u003dTrue, dead_vlan_flow\u003dFalse):"},{"line_number":264,"context_line":"        \"\"\"Plug in the interface only for new devices that don\u0027t exist yet.\"\"\""},{"line_number":265,"context_line":""},{"line_number":266,"context_line":"    def plug(self, network_id, port_id, device_name, mac_address,"}],"source_content_type":"text/x-python","patch_set":12,"id":"b16d5422_93029348","line":263,"range":{"start_line":261,"start_character":4,"end_line":263,"end_character":53},"updated":"2022-01-07 12:04:50.000000000","message":"Be careful: this is overriding the method signature that is used outside Neutron (actually I moved the LinuxInterfaceDriver metaclass definition to n-lib but I didn\u0027t finish the import from Neutron)\n\nTake a look at https://review.opendev.org/c/openstack/neutron/+/821998 and the previous patches.","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"},{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"b07b04fd5342d6539016748d51c19e88f59bcdd2","unresolved":false,"context_lines":[{"line_number":258,"context_line":"        ip_lib.sysctl(cmd, namespace\u003dnamespace)"},{"line_number":259,"context_line":""},{"line_number":260,"context_line":"    @abc.abstractmethod"},{"line_number":261,"context_line":"    def plug_new(self, network_id, port_id, device_name, mac_address,"},{"line_number":262,"context_line":"                 bridge\u003dNone, namespace\u003dNone, prefix\u003dNone, mtu\u003dNone,"},{"line_number":263,"context_line":"                 link_up\u003dTrue, dead_vlan_flow\u003dFalse):"},{"line_number":264,"context_line":"        \"\"\"Plug in the interface only for new devices that don\u0027t exist yet.\"\"\""},{"line_number":265,"context_line":""},{"line_number":266,"context_line":"    def plug(self, network_id, port_id, device_name, mac_address,"}],"source_content_type":"text/x-python","patch_set":12,"id":"0d87ec94_56ca7429","line":263,"range":{"start_line":261,"start_character":4,"end_line":263,"end_character":53},"in_reply_to":"b16d5422_93029348","updated":"2022-01-11 14:43:02.000000000","message":"Thanks for this comment. This got me thinking whether it\u0027s a good ideas to expose such (dhcp- and ovs-agent specific) details in this api at all.\n\nWhile thinking about that, I started to realize this bug is not dhcp specific at all. Yes, we did only see crosstalk between dhcp ports, but l3-agent plugs its ports the exact same way. So it must have the same problems. Because of this the fix should not be dhcp-agent specific either.\n\nSo I\u0027m uploading a new patch set removing the dead_vlan_flow kwarg, but I\u0027m sure that will be breaking a whole lot of tests again...","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"}],"neutron/plugins/ml2/drivers/openvswitch/agent/common/constants.py":[{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"2206570466f4d09671696e2f9eb79fe5d33c526e","unresolved":true,"context_lines":[{"line_number":213,"context_line":""},{"line_number":214,"context_line":"OPENFLOW_MAX_PRIORITY \u003d 65535"},{"line_number":215,"context_line":""},{"line_number":216,"context_line":"# bit marking the presence of 802.1Q header in ovs vlan_tci field"},{"line_number":217,"context_line":"HAS_8021Q_HEADER \u003d 0x1000"},{"line_number":218,"context_line":""},{"line_number":219,"context_line":"# A placeholder for dead vlans."}],"source_content_type":"text/x-python","patch_set":12,"id":"d9dc436d_3eaf73ce","line":216,"updated":"2022-01-07 12:04:50.000000000","message":"There is a constant in os-ken, ofp.OFPVID_PRESENT, that is used in the native implementation","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"},{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"b07b04fd5342d6539016748d51c19e88f59bcdd2","unresolved":false,"context_lines":[{"line_number":213,"context_line":""},{"line_number":214,"context_line":"OPENFLOW_MAX_PRIORITY \u003d 65535"},{"line_number":215,"context_line":""},{"line_number":216,"context_line":"# bit marking the presence of 802.1Q header in ovs vlan_tci field"},{"line_number":217,"context_line":"HAS_8021Q_HEADER \u003d 0x1000"},{"line_number":218,"context_line":""},{"line_number":219,"context_line":"# A placeholder for dead vlans."}],"source_content_type":"text/x-python","patch_set":12,"id":"9daa2c79_49ff6f70","line":216,"in_reply_to":"d9dc436d_3eaf73ce","updated":"2022-01-11 14:43:02.000000000","message":"Done","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"}],"neutron/plugins/ml2/drivers/openvswitch/agent/openflow/native/br_int.py":[{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"2206570466f4d09671696e2f9eb79fe5d33c526e","unresolved":true,"context_lines":[{"line_number":62,"context_line":"            priority\u003dconstants.OPENFLOW_MAX_PRIORITY,"},{"line_number":63,"context_line":"            # The \u0027vlan_vid\u0027 argument of OFPMatch is badly named. Despite the"},{"line_number":64,"context_line":"            # name this value is used as the vlan_tci field in ovs, as written"},{"line_number":65,"context_line":"            # in ovs-fields (7):"},{"line_number":66,"context_line":"            # \"For a packet with an 802.1Q header, this field is the TCI with"},{"line_number":67,"context_line":"            # the bit in CFI’s position (marked P for ``present’’ below)"},{"line_number":68,"context_line":"            # forced to 1.\""}],"source_content_type":"text/x-python","patch_set":12,"id":"68b1a515_7ba159eb","line":65,"range":{"start_line":65,"start_character":13,"end_line":65,"end_character":32},"updated":"2022-01-07 12:04:50.000000000","message":"Right. We already use it in the native implementation of Neutron and os-ken\n        vlan_vid\u003dvlan_tag | ofp.OFPVID_PRESENT\n\nor:\n        if segmentation_id is None:\n            vlan_vid \u003d ofp.OFPVID_NONE\n        else:\n            vlan_vid \u003d segmentation_id | ofp.OFPVID_PRESENT\n\n\nI think this is the only place where we don\u0027t use it. Good catch!\n\nBTW, this is something known in the native implementation. I don\u0027t think we need this comment (just an opinion, nothing else).","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"},{"author":{"_account_id":15554,"name":"Bence Romsics","email":"bence.romsics@gmail.com","username":"ebenrom","status":"working for Ericsson, UTC+1 (+DST)"},"change_message_id":"b07b04fd5342d6539016748d51c19e88f59bcdd2","unresolved":false,"context_lines":[{"line_number":62,"context_line":"            priority\u003dconstants.OPENFLOW_MAX_PRIORITY,"},{"line_number":63,"context_line":"            # The \u0027vlan_vid\u0027 argument of OFPMatch is badly named. Despite the"},{"line_number":64,"context_line":"            # name this value is used as the vlan_tci field in ovs, as written"},{"line_number":65,"context_line":"            # in ovs-fields (7):"},{"line_number":66,"context_line":"            # \"For a packet with an 802.1Q header, this field is the TCI with"},{"line_number":67,"context_line":"            # the bit in CFI’s position (marked P for ``present’’ below)"},{"line_number":68,"context_line":"            # forced to 1.\""}],"source_content_type":"text/x-python","patch_set":12,"id":"183b143a_59fb7f0a","line":65,"range":{"start_line":65,"start_character":13,"end_line":65,"end_character":32},"in_reply_to":"68b1a515_7ba159eb","updated":"2022-01-11 14:43:02.000000000","message":"Done","commit_id":"5d387bd7b3d3c6afac3d104a58e9d5c24edbdf00"}]}