[Openstack-operators] OpenStack-operators Digest, Vol 76, Issue 30

Curtis serverascode at gmail.com
Thu Feb 23 20:50:18 UTC 2017


On Tue, Feb 21, 2017 at 11:14 PM, Y.Rahulan <rahulan.y at gmail.com> wrote:
> Hi,

Hi,

First, a couple house keeping items. You probably just want to join
the operators list, and you probably want to write an email with a
good subject line, and not to reply to a digest email. :)

Now, with that out of the way...

>
>  I am trying to integrate ODL as external OVS manager and network controller
> but I could not mange to connect br-ex as ODL_BRIDGE_MAPPING with public
> interface eth2,
>
> ovs vsctl show:- Manager "tcp:192.67.27.27:6640" is_connected: true Manager
> "ptcp:6641:127.0.0.1" is_connected: true Bridge br-int Controller
> "tcp:192.67.27.27:6653" is_connected: true fail_mode: secure Port br-int
> Interface br-int type: internal Port "tap95af7f98-35" Interface
> "tap95af7f98-35" type: internal Bridge br-ex Port br-ex Interface br-ex
> type: internal Port "eth2" Interface "eth2" ovs_version: "2.5.0"
>
> also it is creating two managers when I create network shown below:-
>
> . openrc admin admin openstack network create admin-external --external
> --provider-network-type flat --provider-physical-network public
> --provider-physical-network physnet1 openstack subnet create ext-subnet
> --network admin-external --allocation-pool
> start=192.67.20.29,end=192.67.20.69 --no-dhcp --gateway 192.67.10.10
> --subnet-range 192.67.27.0/24 openstack router create ext-rtr openstack
> router set ext-rtr --external-gateway admin-external openstack network
> create admin-internal --internal openstack subnet create vx-subnet --network
> admin-internal --subnet-range 192.0.0.0/24 --dns-nameserver 192.67.10.10
> neutron router-interface-add ext-rtr vx-subnet
>
> Is there any one face the same issues and would like to get some answers
> please!
>

What is creating the managers? You shouldn't have two.

Most examples of deploying OpenDaylight show first deploying a
standard OpenStack, deleting all the neutron networks and such, and
wiping out all of the Open vSwitch config, and re-configuring for use
with OpenDaylight. You want to be sure that you have removed
everything properly (assuming a lab deployment that is Ok to mess
around with). You would probably want to write a script to delete
everything while you are testing, so that you can do it over and over
again. :) I had an Ansible playbook that would do that.

The OVS manager is either set manually if a small deployment, or would
usually be part of your configuration management tooling.

That said, there was a bug in Neutron that would reset the manager
[1], and it was patched and backported to Newton Neutron. When I was
testing in my lab I deployed Neutron from source to make sure I had
that patch. This could be part of your issue as well. I'm not sure if
that patch would have made it into various packages repositories yet.
Haven't looked.

So, my guess is that you are not clearing out all the OVS config
properly, and/or are hitting that neutron bug.

Thanks,
Curtis.

[1]: https://review.openstack.org/#/c/403672/



More information about the OpenStack-operators mailing list