Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

I have the following setup: 1 x Management Node (node A) 2 x Compute Nodes (node B & C) 1 x Volume Node (node D)

Relevant configurations: VlanManager multi_host=true

Node B[eth0=192.168.6.102;br100=10.1.0.6] Node C [eth0=192.168.6.103;br100=10.1.0.4]

I can ping between instances on the same node,but not with instances in different nodes. If I run "brctl br100 eth0" , Instances can ping between nodes, but I loose conectivity on eth0 (192.168.6.102/192.168.6.103)

What do I have to change to be able to ping instances between nodes while maintaining normal connectivity on eth0 ?

share|improve this question
    
1. How many physical interfaces on each node? How exactly do they connect, or are the bridged interfaces purely logical interfaces and operate totally within the VM/cloud environment? 2. Do you have a complete config, as in code, (relevant, scrubbed of security blablabla's) for the two nodes? –  C1sc0M4n Nov 13 '12 at 22:04

1 Answer 1

It seems to me that you want to run your nodes with a single physical network interface. What you see is normal, because when you enslave the eth0 physical interface to the br100 bridge, you have to assign IP addresses to the bridge itself, and not to the physical interface.

There is a good example in the OpenStack manual, but it does not cover the case of having your node handle two different IP networks on the same physical interface. You can do that by adding an address to br100. Summing up the relevant parts:

  1. In /etc/nova/nova.conf on Node B, you will have something like:

    my_ip=10.1.0.6
    public_interface=br100
    vlan_interface=eth0
    flat_network_bridge=br100
    flat_interface=eth0
    
  2. In /etc/network/interfaces on Node B, you will have something like:

    auto br100
    iface br100 inet static
        bridge_ports eth0
        bridge_stp off
        bridge_maxwait 0
        bridge_fd 0
        address 192.168.6.102
        netmask 255.255.255.0
        network 192.168.6.0
        broadcast 192.168.6.255
        gateway 192.168.6.xxx
    
    auto br100:1
    iface br100:1 inet static
        address 10.1.0.6
        netmask 255.255.255.0
        broadcast 10.1.0.255
    

I believe that this should work correctly.

share|improve this answer
    
There are actually two interfaces on each node. One is the public interface (eth1) and the other the "internal" one (eth0) Here is the current config of one of the compute nodes: multi_host=true network_manager=nova.network.manager.VlanManager network_host=XPTO vlan_interface=eth0 my-ip=192.168.6.102 enabled_apis=metadata public_interface=eth1 flat_interface=eth0 dmz_cidr=169.254.169.254/32 ec2_dmz_host=169.254.169.254 metadata_host=169.254.169.254 –  drcursor Nov 14 '12 at 8:32

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.