I think we have found (Sylvain and me) a problem that can explain this
trouble:
When the load is too heavy (update dnsmasq host file and send lease update)
on DHCP agent, the report state to Neutron server is delayed and the
Neutron sever considers that agent is down and doesn't sent the port
crea
x27;1.0', u'_context_project_id': None,
> u'_context_timestamp': u'2013-08-08 04:26:09.092921', u'_context_user_id':
> None, u'method': u'update_lease_expiration'}
> 2013-08-08 11:30:16DEBUG [quantum.db.dhcp_rpc_base] Updati
It's not a dnsmasq problem.
I think the problem is: https://bugs.launchpad.net/neutron/+bug/1185916 with
the Neutron Server and the DHCP agent updates loosed when the load is too
high.
Édouard.
On Thu, Aug 22, 2013 at 6:47 AM, Linus Nova wrote:
>
> Hi,
>
> Some times Dnsmasq (dhcp lease offer)
Hi Jon,
Effectively, a bug [1] was identified. When DHCP agent is down because its
overloaded, Neutron agent will not send update to that agent and doesn't
raise the problem.
I submitted small patch to a warning log but I abandoned it to propose
another one (not yet finish) which will raise an exc
Hi,
You can also use instance resource quota [1] to define bandwidth limits on
VNIC.
It's not related to OVS or Neutron. It's Nova specific through the virt
driver libvirt.
[1] https://wiki.openstack.org/wiki/InstanceResourceQuota.
Édouard.
Thanks I will look into this
Remo
Inviato da iPhone
There is also this opening bug:
https://bugs.launchpad.net/neutron/+bug/1185916
I proposed a patch: https://review.openstack.org/#/c/40654/ (I don't know
why it wasn't automatically linked to the launchpad bug! )
I'm waiting feedback/review from core developers.
Regards,
Édouard.
On Wed, Oct
This one
http://fr.slideshare.net/mestery/modular-layer-2-in-openstack-neutron ?
Édouard.
On Wed, Nov 13, 2013 at 11:34 AM, Trinath Somanchi <
trinath.soman...@freescale.com> wrote:
> Hi stackers-
>
>
> Can anyone share the Presentation slides showcased in the Openstack
> Hongkong summit, on
..
>
> Thanks one again..
>
> --
> Trinath Somanchi - B39208
> trinath.soman...@freescale.com | extn: 4048
>
> -Original Message-
> From: Kyle Mestery (kmestery) [mailto:kmest...@cisco.com]
> Sent: Wednesday, November 13, 2013 7:29 PM
> To: Édouard Thuleau
>
Hi all,
I try to set up a plateforme with plugin ML2 and mechanism driver
openvswitch (VXLAN) and l2-pop with a Jumbo frame physical network.
I've got a Cisco Nexus fabric of 10Gb/s and the jumbo frame size is 9216 octets.
I set the MTU on the physical interface used for VXLAN tunneling to 9216 o
Just an update of the subject.
Any ideas ?
Édouard.
On Tue, Dec 3, 2013 at 3:49 PM, Édouard Thuleau
wrote:
> Hi all,
>
> I try to set up a plateforme with plugin ML2 and mechanism driver
> openvswitch (VXLAN) and l2-pop with a Jumbo frame physical network.
> I've got a C
Another alternative is to use the Linux bridge agent with VXLAN tunneling
instead of OVS.
Why you want to use OVS ?
With LB and VXLAN the host topology is clear and understandable.
For example a compute node with 2 VM (a and b) attached to a virtual
network (VNI #1) and another VM (c) attached to
Hi,
I think that script [1] (still under reviewing) will help you.
[1] https://review.openstack.org/#/c/76533/6
Regards,
Édouard.
On Mon, Mar 17, 2014 at 7:41 AM, Damon Wang wrote:
> Hi Simon,
>
> As far as I know, ovs become a driver of ml2, you can use ovs or
> linuxbridge or other drivers
Hi,
I think that script [1] (still under reviewing) will help you.
[1] https://review.openstack.org/#/c/76533/6
Regards,
Édouard.
On Mon, Mar 17, 2014 at 7:41 AM, Damon Wang wrote:
> Hi Simon,
>
> As far as I know, ovs become a driver of ml2, you can use ovs or
> linuxbridge or other drivers
Just point you a blueprint [1] develop during the I release (but deferring
to Juno) to bring L3 HA.
It use VRRP mechanism with keepalived that you could reuse for LBaaS.
It also uses conntrackd to synchronize connections between routers (master
and slave). I don't know how to configure LB synchroni
14 matches
Mail list logo
