Ask Your Question

Revision history [back]

I think I am having the same problem.

I'll try to explain it on a different way:

NETWORK NODE:

 [root@network ~]# ip netns
 qrouter-02ec0122-0fe7-4f6e-bb90-e00655fc9f22
 qdhcp-daf16c13-c18e-4ad7-b0a6-cf625da81b5d

[root@network ~]# ip netns exec qdhcp-daf16c13-c18e-4ad7-b0a6-cf625da81b5d ip a  |grep tap
9: tapd83ea76c-c3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN 
    inet 192.168.1.3/24 brd 192.168.1.255 scope global tapd83ea76c-c3

COMPUTE NODE:

[root@compute1 ~]# virsh dumpxml instance-00000009 |grep tap
      <target dev='tapc7e97925-58'/>

How can it be that the Test Access Point (TAP) ID of the interface generated by neutron is different of the one used by the instance?

Consequently, as in the case of Luis, the VM tries to connect to a wrong interface, and does not get any IP from the DHCP.

Does anybody know how to solve this?

(My set up is exactly as the recommended one in the documentation: CentOS 7, Juno, 5 VMs ==> controller, network, compute, block, object)