[qeeens][neutron] migrating from iptables_hybrid to openvswitch

Jakub Libosvar jlibosva at redhat.com
Thu Mar 19 12:54:12 UTC 2020

On 13/03/2020 08:24, Ignazio Cassano wrote:
> Hu Jakub, migrating vm from a not with hybrid_itatabes ti a node switched
> on openvswitch works fine . The problem is this migration create the qbr on
> the mode switched to openvswitch.
> But when I switch another compute node to openvswitch and I try to live
> migrate the same vm (openvswitch to qopenswitch) it does not work because
> the qbr presence.
> I verified on nova logs.
> Ignazio

Hi Ignazio,

I think the first step - migrating from hybrid_iptables to ovs should
not create the qbr on the target node. It sounds like a bug - IIRC the
libvirt domxml should not have the qbr when migrating.

> Il Gio 12 Mar 2020, 23:15 Jakub Libosvar <jlibosva at redhat.com> ha scritto:
>> On 12/03/2020 11:38, Ignazio Cassano wrote:
>>> Hello All, I am facing some problems migrating from iptables_hybrid
>>> frirewall to openvswitch firewall on centos 7 queens,
>>> I am doing this because I want enable security groups logs which require
>>> openvswitch firewall.
>>> I would like to migrate without restarting my instances.
>>> I startded moving all instances from compute node 1.
>>> Then I configured openvswitch firewall on compute node 1,
>>> Instances migrated from compute node 2 to compute node 1 without
>> problems.
>>> Once the compute node 2 was empty, I migrated it to openvswitch.
>>> But now instances does not migrate from node 1 to node 2 because it
>>> requires the presence of qbr bridge on node 2
>>> This happened because migrating instances from node2 with iptables_hybrid
>>> to compute node 1 with openvswitch, does not put the tap under br-int as
>>> requested by  openvswich firewall, but qbr is still present on compute
>> node
>>> 1.
>>> Once I enabled openvswitch on compute node 2, migration from compute
>> node 1
>>> fails because it exprects qbr on compute node 2 .
>>> So I think I should moving on the fly tap interfaces from qbr to br-int
>> on
>>> compute node 1 before migrating to compute node 2 but it is a huge work
>> on
>>> a lot of instances.
>>> Any workaround, please ?
>>> Ignazio
>> I may be a little outdated here but to the best of my knowledge there
>> are two ways how to migrate from iptables to openvswitch.
>> 1) If you don't mind the intermediate linux bridge and you care about
>> logs, you can just change the config file on compute node to start using
>> openvswitch firewall and restart the ovs agent. That should trigger a
>> mechanism that deletes iptables rules and starts using openflow rules.
>> It will leave the intermediate bridge there but except the extra hop in
>> networking stack, it doesn't mind.
>> 2) With multiple-port binding feature, what you described above should
>> be working. I know Miguel spent some time working on that so perhaps he
>> has more information about which release it should be functional at, I
>> think it was Queens. Not sure if any Nova work was required to make it
>> work.
>> Hope that helps.
>> Kuba

More information about the openstack-discuss mailing list