Open Stack

Hello,

i have a setup with a Controller node and two Compute nodes. The topology
is as in the figure attached. I followed the hastexo manual <
http://www.hastexo.com/resources/docs/installing-openstack-essex-20121-ubuntu-1204-precise-pangolin>,
but installing just the modules as described in the figure (attached).
The private network is the nova fixed_range=192.168.22.32/27
The public network is the nova floating_range=10.4.5.32/27
I can launch VMs and access them with no problem.
But from a VM inside the cloud I cannot access any machine from a range
other than the fixed_range (192.168.22.32/27).

Does anyone know what can it be?
How can the controller resolve NAT coming from a node outside the cloud,
while cannot resolve coming from a VM in the cloud?


Also, i would like to understand the nova-network setup using flat dhcp.

br100 in controller assumes the ip 192.168.22.33 (from dhcp i guess,
different from eth1 that is 192.168.22.1).
In the computes, br100 assumes the same ip as the eth1.
All nodes have a vrbr0 with the same ip 192.168.122.1, why '.122.'?


br100 bridges all the VMs inside a physical server and vrbr0 bridges all
br100, is that?

# network specific settings of nova.conf
--network_manager=nova.network.manager.FlatDHCPManager
--public_interface=eth0
--flat_interface=eth1
--flat_network_bridge=br100
--fixed_range=192.168.22.32/27
--floating_range=10.4.5.32/27
--network_size=32
--flat_network_dhcp_start=192.168.22.33
--flat_injected=False
--force_dhcp_release

Ah, i actually do not know if this is the best topology for the setup.

I appreciate any help.

Thanks.

-- 
*Alisson*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20120518/51759f9a/attachment.html>