Not able to ping gateway

Asked by David Thomas

Hi,
I have succesfully set up openstack in ubuntu 13.10(server edition). I have done all in one(AIO) i.e.., installed all services in single node. Now, i created instances using dashboard and attached them to a particular network. But the problem is instances are not able to ping gateway of that network. Instances are able to ping between themselves. I got struck at this point. There may be any problem ovs-plugin configurations. Can anyone help me regarding this??

Thanks
Naga Phani

Question information

Language:
English Edit question
Status:
Solved
For:
neutron Edit question
Assignee:
No assignee Edit question
Solved by:
Vish Ishaya
Solved:
Last query:
Last reply:
Revision history for this message
Brian Lamar (blamar) said :
#1

This is not the case, although I can't help you with your specific case, I have run a cluster with both Libvirt and XenAPI hypervisors. A single compute instance can only control one type of hypervisor, but there should be able to be multiple types in a cluster.

Revision history for this message
Best Vish Ishaya (vishvananda) said :
#2

This was a bug in cactus which has been fixed in trunk

https://bugs.launchpad.net/nova/+bug/787657

Vish

On Jun 15, 2011, at 10:51 AM, David Thomas wrote:

> New question #161562 on OpenStack Compute (nova):
> https://answers.launchpad.net/nova/+question/161562
>
> I've successfully set up OpenStack on Ubuntu 11.04 with a single controller node and multiple compute nodes using KVM as the hypervisor. I have an ESXi 4.1 machine that I wanted to add to that cluster. So I modified the nova.conf file on one of the compute nodes to manage the ESXi server instead. I followed the VMware ESX/ESXi Server Support for OpenStack Compute document (http://nova.openstack.org/vmwareapi_readme.html) for setting up the compute node. After restarting the compute node, I got fatal errors in the nova-compute.log saying that the vlan_interface flag had two different values. It appears the compute node is seeing the flag in it's own nova.conf and the one in the controller node's nova.conf. This leads me to believe that a single OpenStack cluster can only have one type of hypervisor. I thought I may have seen that in some documentation somewhere but now I can't find it to confirm. Does anyone know if this is the case?
>
> Thanks
> Dave Thomas
>
>
> --
> You received this question notification because you are a member of Nova
> Core, which is an answer contact for OpenStack Compute (nova).

Revision history for this message
David Thomas (david-thomas-commitent) said :
#3

Thanks Vish Ishaya, that solved my question.

Revision history for this message
David Thomas (david-thomas-commitent) said :
#4

Thank you both for the help.