Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-02-03 Thread Jonathan Proulx
x27;m completely wrong here. > > Marek > > -Original Message- > From: Jonathan Proulx [mailto:j...@jonproulx.com] > Sent: 30. januára 2014 19:11 > To: Robert Collins > Cc: openstack@lists.openstack.org > Subject: Re: [Openstack] [Neutron] asymetric DHCP brokenne

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-31 Thread Ruzicka, Marek
shooting from the hip here, so sorry if I'm completely wrong here. Marek -Original Message- From: Jonathan Proulx [mailto:j...@jonproulx.com] Sent: 30. januára 2014 19:11 To: Robert Collins Cc: openstack@lists.openstack.org Subject: Re: [Openstack] [Neutron] asymetric DHCP brokenne

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-30 Thread Jonathan Proulx
Still can't quite sort this out but I am circling in on where the problem is. To recap bootpc and arp requests from instances using GRE tenant networks are not making it onto the physical network, I suspect this is "all broadcast traffic". If IP is configured statically and the arp cache is set

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Jonathan Proulx
On Wed, Jan 29, 2014 at 3:39 PM, Robert Collins wrote: > On 30 January 2014 08:16, Jonathan Proulx wrote: >> On Wed, Jan 29, 2014 at 1:49 PM, Joe Topjian wrote: > Always use ovs-vsctl show on ovs switches - brcompat is super limited. usually do just interesting that what looks the same in ovs-

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Robert Collins
On 30 January 2014 08:16, Jonathan Proulx wrote: > On Wed, Jan 29, 2014 at 1:49 PM, Joe Topjian wrote: >> >>> however I can't tcpdump on the patch or gre devices >>> >>> # tcpdump -i patch-tun >>> tcpdump: patch-tun: No such device exists >> >> >> I can reproduce this. I suspect because patch

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Joe Topjian
> This may be getting close to the issue. I don't see any interfaces > anything like that. I'm seeing two different types of bride states on > my compute nodes, which suggest something's wrong there. On the > compute node hosting the 'bad' instances and many other nodes as well > I see: > > bridg

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Jonathan Proulx
On Wed, Jan 29, 2014 at 1:49 PM, Joe Topjian wrote: > >> however I can't tcpdump on the patch or gre devices >> >> # tcpdump -i patch-tun >> tcpdump: patch-tun: No such device exists > > > I can reproduce this. I suspect because patch-tun and patch-int are OVS > patch interfaces, they are inte

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Joe Topjian
> however I can't tcpdump on the patch or gre devices > > # tcpdump -i patch-tun > tcpdump: patch-tun: No such device exists > I can reproduce this. I suspect because patch-tun and patch-int are OVS patch interfaces, they are internal to OVS and not a real interface. "ip a | grep patch-tun" re

Re: [Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Jonathan Proulx
I also notice ARPs are asysmetric. arp requests from the instance don't end up on the wire, but arp requests from the network controller do make it to the instance and the responses make it back. So once the network controller has made contact and it's MAC is in the instance ARP cache then things

[Openstack] [Neutron] asymetric DHCP brokenness on tenant GRE networks

2014-01-29 Thread Jonathan Proulx
HI All, DHCP requests from instances with interfaces on OVS/GRE based tenant networks are showing up on the tap device on the compute node but never make it to the physical network device (tcpdump -i ehtX proto gre). If I manually configure an address all seems well & I can for example ping from