[ovs-discuss] tun_src prefixes of flow_tables not work?

wenxu wenx05124561 at 163.com
Sat Oct 14 02:37:36 UTC 2017


Hi all,


I just test the prefixes function in flow_tables, It seems not work for tun_src


# ovs-vsctl list flow_table
_uuid               : 113ce6a4-f872-450b-b799-e15f3f3af7c1
external_ids        : {}
flow_limit          : []
groups              : []
name                : "table0"
overflow_policy     : []
prefixes            : [ip_dst, ip_src, tun_src]


# ovs-vsctl list bridge test-br
_uuid               : 63ec1932-bf2d-441a-bcc1-d730a47f0156
auto_attach         : []
controller          : []
datapath_id         : "00003219ec631a44"
datapath_type       : ""
datapath_version    : "2.6.0"
external_ids        : {}
fail_mode           : []
flood_vlans         : []
flow_tables         : {0=113ce6a4-f872-450b-b799-e15f3f3af7c1}
ipfix               : []
mcast_snooping_enable: false
mirrors             : []
name                : test-br
netflow             : []
other_config        : {}
ports               : [961cd756-c8ce-41ab-b208-67aad998333d, fe22c308-73f9-4c44-a2dc-529f5b7a0dd1]
protocols           : []
rstp_enable         : false
rstp_status         : {}
sflow               : []
status              : {}
stp_enable          : false




with flows: ofport 1 is tunnel port

# ovs-ofctl dump-flows test-br in_port=1

'tun_src=10.0.0.0/24,in_port=1 actions=LOCAL'


but the tunnel src still unwildcard in datapth flow 



recirc_id(0),tunnel(src=10.0.0.2,dst=10.0.0.1,flags(-df-csum)),in_port(2),skb_mark(0),eth_type(0x0800),ipv4(frag=no), packets:27, bytes:2646, used:0.052s, actions:5
recirc_id(0),tunnel(src=10.0.0.3,dst=10.0.0.1,flags(-df-csum)),in_port(2),skb_mark(0),eth_type(0x0800),ipv4(frag=no), packets:14, bytes:1372, used:0.629s, actions:5


Are there some problem for my configuration?






BR
wenxu




【网易自营】好吃到爆!鲜香弹滑加热即食,经典13香/麻辣小龙虾仅75元3斤>>      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openvswitch.org/pipermail/ovs-discuss/attachments/20171014/1c4b48e4/attachment-0001.html>


More information about the discuss mailing list