Ask Your Question
0

neutron - Not able to ping virtual machine from host

asked 2014-01-06 05:10:24 -0500

thota gravatar image

I have a requirement from a customer (telecom application) that - Eight virtual machines are launched on a compute node using openstack nova and each virtual machine shall run the telecom application. The telecom application running on the virtual machine handles different types of traffic (for ex, management and data traffics). These traffics are differentiated with VLAN IDs. The compute node has a 10Gbps NIC that receives the tagged data.

With this, I want to create a OVS that shall trunk the tagged traffic into the virtual machine. That is, OVS should not strip the vlan tag and the application on the VM should see the vlan tag to differentiate the traffic.

Now, to start with, I want to establish communication between host (compute) and VM using OVS. In simple, I should be able to ping the virtual machine from the host (compute1) through OVS. For this, I created OVS bridge "br-int" and created a network using openstack neutron and launched a virtual machine. openstack nova created a tap interface (tap63ea2815-b5) that got connected to eth0 in the virtual machine and to OVS on the compute node. Later I added eth11 interface (10Gbps NIC on compute/host) to the ovs bridge.

Configured both eth11 (on the host/compute) and eth0 on the virtual machine in the same subnet. I’m not able to ping the VM from the host.

But, If I remove the IP address on eth11 and assign it to br-int, then I’m able to ping the virtual machine.

Can someone explain the behavior.

Below is “ovs-vsctl show” output [root@compute1 ~]# ovs-vsctl show 0bfed3e4-5f13-4726-a6a5-073fd47f1500 Bridge br-int Port "eth11" Interface "eth11" Port "tapb62b2a46-71" Interface "tapb62b2a46-71" Port br-int Interface br-int type: internal ovs_version: "1.11.0"

Thanks in advance.

edit retag flag offensive close merge delete

1 answer

Sort by » oldest newest most voted
0

answered 2014-01-06 05:21:36 -0500

thota gravatar image

I got the information from openvswitch FAQ. "There is no compelling reason why Open vSwitch must work this way.However, this is the way that the Linux kernel bridge module has always worked, so it's a model that those accustomed to Linux bridging are already used to. Also, the model that most people expect is not implementable without kernel changes on all the versions of Linux that Open vSwitch supports."

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2014-01-06 05:10:24 -0500

Seen: 1,035 times

Last updated: Jan 06 '14