Ask Your Question

dsmith4546's profile - activity

2018-10-05 05:23:51 -0600 received badge  Famous Question (source)
2015-10-01 04:38:03 -0600 received badge  Notable Question (source)
2015-05-04 09:50:29 -0600 received badge  Popular Question (source)
2015-03-02 12:23:46 -0600 received badge  Famous Question (source)
2014-10-03 09:53:26 -0600 received badge  Notable Question (source)
2014-09-03 04:04:31 -0600 received badge  Popular Question (source)
2014-06-30 21:40:10 -0600 received badge  Famous Question (source)
2014-05-22 07:53:31 -0600 received badge  Notable Question (source)
2014-05-22 07:53:31 -0600 received badge  Popular Question (source)
2013-06-26 16:50:04 -0600 answered a question Creating a VM to act as a router

I think I need to start digging into the quantum code more. I will ask some other folks I work with to see if they are interested in adding that functionality.

2013-06-26 16:48:49 -0600 answered a question Creating a VM to act as a router

Thanks Aaron Rosen, that solved my question.

2013-06-17 17:41:31 -0600 answered a question Creating a VM to act as a router

I am 99% sure this is has to do with OpenStack not allowing IP spoofing. Is there a way to turn off IP spoofing in Grizzly?

2013-06-15 16:37:38 -0600 answered a question Creating a VM to act as a router

Yes I have enabled IP forwarding. I have an OpenVPN tunnel between two vms in two different clouds. Each of those VMs are doing routing for their local subnet they are connected to. I have the appropriate routes in place on all virtual machines. I want to ping from a VM on one cloud to a VM on another cloud. I want to go over the OpenVPN tunnel. I am doing this as you said above by using those OpenVPN vms as the gateway. I see the traffic going over the tunnel. I see the traffic hitting the far side VM. The strange thing is I see the echo reply coming back. But it STOPS at the VM that is the gateway for my VM I'm pinging from. When I run a capture on the actual ethernet interface for that VM that is running OpenVPN, and is doing the routing, I see echo replies.....but they are never routed back to the VM who sent out the echo requests. I ran a capture on the tap interface and am seeing echo replies, but when I run a capture the q interface on ovs I don't see the echo replies..... Normally I was going to say there must be some kind of quantum rule for IP spoofing or something, but that doesn't make any sense because the traffic gets all the way to the far side VM over the tunnel. And when it gets routed at the other side of the tunnel the source address is an address that is different then the IP address assigned to the interface. This probably isn't an OpenStack issue.

2013-06-13 17:51:54 -0600 asked a question Creating a VM to act as a router

Hello,

I recently setup a virtual machine to act as a router. The problem is the vm is not routing traffic correctly. This is essentially what my setup is, refer to the output below.

(VM IM PINGING FROM) (Private subnet X) | (OPEN VPN TUNNEL) | (VMs below is in Private subnet Y on a completely different cloud) (VM ACTING AS ROUTER)(INT-X)--(INT-Y)(OVS)--------(VM IM TRYING TO PING)

When I run a packet capture on INT-X on the router VM. I see echo requests going out. When I run a packet capture on INT-Y on the OVS. I no longer see the echo requests........

Is there some sort of libvirt rule that needs to be changed in order for a virtual machine to act as a router?

2013-05-16 14:00:57 -0600 asked a question GRE Key for virtual networks using OVS plugin

hello,

I have asked this question before, but what I am trying to do is still get a basic understanding of the way the quantum server, atleast I think it's the quantum server...... exchanges information with the ovs agents running on the compute nodes. I have installed grizzly, I have 3 compute nodes, and one network/controller node. I I ran some packet captures and noticed that there is a GRE Key value associated with each virtual network that I created within the tenants. The key is listed as 0x000000#. I then looked in the quantum database at the ovs_network_bindings table, and the column segmentation_id. I noticed that for each network_id in that table, the segmentation_id matched the network_id of the network that I have created thus far. I guess my question is how does ovs_agent on the compute node, know to tag the traffic with a specific key value for a specific network? How does this key value get relayed from the mysql database down to the ovs agents? Is there any documentation/graphic that expresses how this information exchange works? Any information would be appreciated.

2013-03-28 12:36:09 -0600 answered a question How does access to the metadata service work when using quantum?

As a reminder we had to add routes on the host that is running the metadata service that point back to the ip address of the l3 agent. For each private network that you create.

Example)

route add -net 192.168.4.0/24 gw 192.168.38.199 <---That .199 is the address of the l3 agent. route add -net 192.168.5.0/24 gw 192.168.38.199

2013-03-28 12:24:56 -0600 answered a question How does access to the metadata service work when using quantum?

Hey, looks like we fixed it, without any custom iptables rules. We followed this documented here: http://docs.openstack.org/folsom/openstack-network/admin/content/adv_cfg_l3_agent_metadata.html (http://docs.openstack.org/folsom/open...)

The page stated that the nova-metadata service must be reachable from the external network. Sense this is the case I updated the metadata ip in the nova.conf and the l3-agent.ini files to reflect the external ip address of the box that is running the metadata service. So in our case we were running the metadata service on the controller. After that, we restarted all of the nova-services and quantum services. We are now able to curl the following address from our vms: 'curl http://169.254.169.254/openstack'

2013-03-27 19:42:47 -0600 asked a question How does access to the metadata service work when using quantum?

Hello,

I used this install guide https://github.com/mseknibilel/OpenStack-Folsom-Install-guide/blob/stable/GRE/OpenStack_Folsom_Install_Guide_WebVersion.rst (https://github.com/mseknibilel/OpenSt...) . We are not using namespaces.

We have the metadata service installed on the controller, howcome I can't see the 169.254.169.254 address listed anywhere anywhere on the controller. Where does that address live?

If the VM needs to talk to the 169.254.169.254 address, will the layer 3 agent change it to whatever address you have listed for the matadata_ip in the l3_agent.ini file?

2013-03-26 12:37:26 -0600 answered a question Grizzly Dashboard and Quantum

Thanks Jason, that solved my question.

2013-03-25 21:23:46 -0600 asked a question Grizzly Dashboard and Quantum

Alright here is the list of things we cannot do from within the Folsom dashboard. I just want to check to see if these features will be available in the Grizzly dashboard.

Will one be able to create a router within a tenant and connect subnets to that router within a tenant. - Answered

Will one be able to assign floating ips to instances from within the grizzly dashboard?

Will one be able to assign a dns-server or servers to a specific subnet from within the dashboard?

Will one be able to change dns-nameservers for subnets that have already been created, from within the dashboard?

Will one be able to create a subnet that is shared among all tenants from within the dashboard?

Not related to the dashboard Will one have the ability to use namespaces, and have support for the metadata service and security groups for individual tenants?

2013-03-13 13:51:13 -0600 answered a question Will Grizzly give one the ability to create a layer 3 agent from the dashboard?

Thanks dan wendlandt, that solved my question.

2013-03-12 14:41:16 -0600 asked a question Will Grizzly give one the ability to create a layer 3 agent from the dashboard?

Hello,

Will the dashboard in grizzly have the ability to allow one to create a layer 3 agent or agents for a particular tenant?

Any information would be appreciated.

Thanks, doug

2013-02-15 19:38:58 -0600 answered a question GRE Tunnels and VLANs on OVS

I guess I need to understand ovs a little better. How do two ovs's connected together with a gre tunnel know what to do with a broadcast frame from a particular vlan? How does the ovs make a layer 2 decision if the frame isn't tagged with a vlan id?

2013-02-14 12:57:59 -0600 answered a question GRE Tunnels and VLANs on OVS

Thanks GMi, that solved my question.

2013-02-09 20:41:29 -0600 asked a question GRE Tunnels and VLANs on OVS

Very simple question. I used this guide to install folsom: http://docs.openstack.org/folsom/basic-install/content/basic-install_intro.html (http://docs.openstack.org/folsom/basi...) . I am not using namespaces. I am using the gre tenant_network_type. The interface that dnsmasq hands out ip addresses on, and the gateway interface that the vm uses to leave the subnet, are in completely different vlans than any of the vm interfaces on the ovs on the compute node. I made an entry here: http://visualne.wordpress.com/2013/02/09/quantum-gre-tunnels-ovs/ (http://visualne.wordpress.com/2013/02...) . I guess my question is, do the gateway and dnsmasq interface need to be in the same vlan as the vm interface?

2012-12-18 01:43:23 -0600 answered a question Gateway addresses for subnets are one off.

So all I had to do was add the subnet to the router and it worked. So the first thing I did was do a "quantum subnet-list", I noticed that the subnet I made was in the list. I then did a "quantum port-list -- --device_id=76930773-9c65-4742-9e43-2efccbd66d7f" and noticed the subnet I created wasn't "connected" to that router. So then I did a "quantum router-interface-add 76930773-9c65-4742-9e43-2efccbd66d7f 67f4fa58-b293-439f-961a-8b892d4313f6" And that created the gateway address on the router. Like I said before though for whatever reason when I create the subnet in the dashboard, it won't attach the subnet I created to the router for that tenant, and obviously if it won't do that then no gateway for that subnet gets created. That might be something specific to my config or a bug, not really sure.

############ [DEFAULT]

Show debugging output in log (sets DEBUG log level output)

debug = True

L3 requires that an interface driver be set. Choose the one that best

matches your plugin.

OVS

interface_driver = quantum.agent.linux.interface.OVSInterfaceDriver

LinuxBridge

#interface_driver = quantum.agent.linux.interface.BridgeInterfaceDriver

The Quantum user information for accessing the Quantum API.

auth_url = http://192.168.1.2:35357/v2.0 auth_region = RegionOne admin_tenant_name = service admin_user = quantum admin_password = test123

Use "sudo quantum-rootwrap /etc/quantum/rootwrap.conf" to use the real

root filter facility.

Change to "sudo" to skip the filtering and just run the comand directly

root_helper = sudo quantum-rootwrap /etc/quantum/rootwrap.conf #root_helper = sudo /usr/bin/quantum-rootwrap /etc/quantum/rootwrap.conf

Allow overlapping IP (Must have kernel build with CONFIG_NET_NS=y and

iproute2 package that supports namespaces).

use_namespaces = False

If use_namespaces is set as False then the agent can only configure one router.

This is done by setting the specific router_id.

router_id = 76930773-9c65-4742-9e43-2efccbd66d7f

Each L3 agent can be associated with at most one external network. This

value should be set to the UUID of that external network. If empty,

the agent will enforce that only a single external networks exists and

use that external network id

gateway_external_net_id = a25c67f6-cfd7-4f97-b550-450ceac90b7a

Indicates that this L3 agent should also handle routers that do not have

an external network gateway configured. This option should be True only

for a single agent in a Quantum deployment, and may be False for all agents

if all routers must have an external network gateway

handle_internal_only_routers = True

Name of bridge used for external network traffic. This should be set to

empty value for the linux bridge

external_network_bridge = br-ex

IP address used by Nova metadata server

metadata_ip = 192.168.1.2

TCP Port used by Nova metadata server

metadata_port = 8775

The time in seconds between state poll requests

polling_interval = 3

2012-12-17 03:47:51 -0600 answered a question Gateway addresses for subnets are one off.

I got it. For whatever reason when I go to create a new network within the tenant, a gateway isn't created on the layer3 agent. I have to go in and manually add it to the layer 3 agent. Good stuff though. Thanks for the help.

2012-12-16 22:07:43 -0600 answered a question Gateway addresses for subnets are one off.

Alright I got it working now. I see the gateway for one of the private subnets from my vm just fine. The only problem is when I try to create another network through the dashboard. The dashboard creates a dhcp server to hand out ips for that subnet but a gateway won't be created for that subnet on the layer 3 agent. I am not using namespaces. Is it possible to not use namespaces and have more then one private network off of a layer 3 agent?

2012-12-14 20:41:37 -0600 answered a question Gateway addresses for subnets are one off.

Alright I have some more information, below is a piece of the /etc/quantum/l3_agent.ini file.

Allow overlapping IP (Must have kernel build with CONFIG_NET_NS=y and

iproute2 package that supports namespaces).

use_namespaces = False

If use_namespaces is set as False then the agent can only configure one router.

This is done by setting the specific router_id.

router_id = 296ccc78-1766-41c9-9ead-713a0843070f

I checked the internal ports on router_id "296ccc78-1766-41c9-9ead-713a0843070f" with

quantum port-list -- --device_id=296ccc78-1766-41c9-9ead-713a0843070f

And I saw both the internal port with ip of 10.6.6.1 and the external port with ip 192.168.2.130

Any idea why I'm still not seeing any 10.6.6.1 ip address with ifconfig -a? And yes, you are right I am not using namespaces.

2012-12-13 00:59:48 -0600 answered a question Gateway addresses for subnets are one off.

I'm running Ubuntu 12.04.1 LTS kernel 3.2.0-33 generic - Running on box where l3 agent, dhcp agent, ovs agent is running I'm running Ubuntu 12.04.1 LTS kernel 3.2.0-29-generic - Running on compute node where ovs agent is running

When I run "ip netns" on both the compute node and box where l3 agent, dhcp agent, ovs agent is running, I dont see anything. In googling around I believe I saw at one point the 3.2.0-33 kernel was supported. The 29 one I am going look into now. Am I not seeing any output from ip netns because network namespaces aren't supported?

2012-12-11 21:41:28 -0600 answered a question Gateway addresses for subnets are one off.

Alright. I understand. The problem is, when my vm wants to leave the subnet, it will send an arp out for the gateway address of .1. The arp will go over the gre tunnel to the l3 agent. But the gateway ip address of .1 does not exist on the layer 3 agent, so there will be no response to the arp. If I issue the command "ifconfig -a" on the box that is running the layer 3 agent I don't see a .1 address anywhere. So my question is this, if I create a gateway address for a tenant subnet via the dashboard. Will it actually get created anywhere...

2012-12-11 12:04:00 -0600 answered a question Gateway addresses for subnets are one off.

Ok, so the .2 address is handing out IPs. Ok. Thats fine. So why does the dashboard prompt me for a gateway address? Do I have to create a gateway address myself?

2012-12-10 00:36:01 -0600 answered a question Gateway addresses for subnets are one off.

Ya, I understand that, but why is an ip address being created as .2 and not .1?

2012-12-09 23:59:35 -0600 answered a question Gateway addresses for subnets are one off.

I am running folsom on ubuntu 12.04. I used this guide. http://docs.openstack.org/folsom/basic-install/content/basic-install_intro.html (http://docs.openstack.org/folsom/basi...)

2012-12-09 23:58:31 -0600 asked a question Gateway addresses for subnets are one off.

Here are the two private subnets I have created thus far for a tenant.

192.168.4.0/24 GW: 192.168.4.1 10.6.6.0/24 GW: 10.6.6.1

I am using the ovs agent. I configured the gateways for each of the subnets to be .1 addresses. But when I do an ifconfig on the same box my quantum-dhcp-agent, ovs-agent, l3-agent are running on, I see all .2 addresses for those subnets…. Refer to the output below.

tap882a8e75-ee Link encap:Ethernet HWaddr 3e:11:25:db:30:ed
inet addr:10.6.6.2 Bcast:10.6.6.255 Mask:255.255.255.0 inet6 addr: fe80::3c11:25ff:fedb:30ed/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:74550 errors:0 dropped:0 overruns:0 frame:0 TX packets:35835 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:13741301 (13.7 MB) TX bytes:6537014 (6.5 MB)

tap3680cb06-ab Link encap:Ethernet HWaddr fa:16:3e:2e:eb:6b
inet addr:192.168.4.2 Bcast:192.168.4.255 Mask:255.255.255.0 inet6 addr: fe80::f816:3eff:fe2e:eb6b/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:4943 errors:0 dropped:0 overruns:0 frame:0 TX packets:3132 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:1004139 (1.0 MB) TX bytes:550835 (550.8 KB)

VMs are getting IP addresses fine, but they are being assigned a gateway address that don't exist anywhere. My VMs that are spun up are getting .3 addresses. Any input would be appreciated.

2012-12-05 02:00:47 -0600 answered a question Can't ping gateway of tenant router....

Thanks yong sheng gong, that solved my question.

2012-12-05 02:00:33 -0600 answered a question Can't ping gateway of tenant router....

I got it. Final ovs configuration listed below. Thanks for the help.

Controller


root@controller:~# ovs-vsctl show 195ff79e-e01a-4e93-84bc-868d8782f284 Bridge br-ex Port "qg-2b9a29bb-4b" Interface "qg-2b9a29bb-4b" type: internal Port "eth2" Interface "eth2" Port br-ex Interface br-ex type: internal Bridge br-tun Port br-tun Interface br-tun type: internal Port patch-int Interface patch-int type: patch options: {peer=patch-tun} Bridge br-int Port "gre-1" Interface "gre-1" type: gre options: {remote_ip="10.1.1.2"} Port "tap8320bbe5-23" tag: 2 Interface "tap8320bbe5-23" type: internal Port patch-tun Interface patch-tun type: patch options: {peer=patch-int} Port "tap882a8e75-ee" tag: 1 Interface "tap882a8e75-ee" type: internal Port "tap4265fa27-08" tag: 4095 Interface "tap4265fa27-08" type: internal Port br-int Interface br-int type: internal Port "qr-1549a07f-3a" tag: 4095 Interface "qr-1549a07f-3a" type: internal ovs_version: "1.4.0+build0"

Compute Node


root@ComputeNode1:~# ovs-vsctl show 331fd29b-b70f-4f54-983a-d6749f8cf1ed Bridge br-int Port patch-tun Interface patch-tun type: patch options: {peer=patch-int} Port "qvo235c2e1a-dd" tag: 1 Interface "qvo235c2e1a-dd" Port "qvo002cc4c6-47" tag: 1 Interface "qvo002cc4c6-47" Port "qvo984554c8-98" tag: 3 Interface "qvo984554c8-98" Port br-int Interface br-int type: internal Bridge br-tun Port "gre-1" Interface "gre-1" type: gre options: {in_key=flow, out_key=flow, remote_ip="10.1.1.1"} Port patch-int Interface patch-int type: patch options: {peer=patch-tun} Port br-tun Interface br-tun type: internal ovs_version: "1.4.0+build0"

2012-12-05 00:47:45 -0600 answered a question Can't ping gateway of tenant router....

I am running ubuntu 12.04, and I followed this guide for installing folsom http://docs.openstack.org/folsom/basic-install/content/basic-install_intro.html (http://docs.openstack.org/folsom/basi...)

2012-12-04 15:41:47 -0600 answered a question Can't ping gateway of tenant router....

Anyone have any thoughts on this one?

2012-12-02 17:06:46 -0600 answered a question Can't ping gateway of tenant router....

Sorry problem still exists. Good news is it looks like the arp is trying to go over the gre tunnel. The problem is, the gre endpoint is not responding to the arp request. So here's what I have

Controller - Private IP - eth1 - 10.1.1.1 Compute Node - Private IP - eth1 - 10.1.1.2

OVS Configuration - Controller root@controller:~# ovs-vsctl show 195ff79e-e01a-4e93-84bc-868d8782f284 Bridge br-ex Port "qg-2b9a29bb-4b" Interface "qg-2b9a29bb-4b" type: internal Port "eth2" Interface "eth2" Port br-ex Interface br-ex type: internal Bridge br-tun Port br-tun Interface br-tun type: internal Port patch-int Interface patch-int type: patch options: {peer=patch-tun} Bridge br-int Port patch-tun Interface patch-tun type: patch options: {peer=patch-int} Port "tap882a8e75-ee" tag: 1 Interface "tap882a8e75-ee" type: internal Port "tap4265fa27-08" tag: 4095 Interface "tap4265fa27-08" type: internal Port "eth1" Interface "eth1" Port br-int Interface br-int type: internal Port "qr-1549a07f-3a" tag: 4095 Interface "qr-1549a07f-3a" type: internal ovs_version: "1.4.0+build0"

OVS Configuration - ComputeNode1 root@ComputeNode1:~# ovs-vsctl show 331fd29b-b70f-4f54-983a-d6749f8cf1ed Bridge br-int Port patch-tun Interface patch-tun type: patch options: {peer=patch-int} Port "qvo002cc4c6-47" tag: 1 Interface "qvo002cc4c6-47" Port br-int Interface br-int type: internal Bridge br-tun Port "gre-1" Interface "gre-1" type: gre options: {in_key=flow, out_key=flow, remote_ip="10.1.1.1"} Port patch-int Interface patch-int type: patch options: {peer=patch-tun} Port br-tun Interface br-tun type: internal ovs_version: "1.4.0+build0"

When I ping from the vm private ip of 10.5.5.3 to the gateway I have setup that lives on the l3 agent on the controller. Eth1 on the compute node will arp for eth1s ip on the controller. But eth1 on the controller isn't responding. Refer to the output below. I am seeing a gre port configured on the compute node but I am not seeing one configured on the controller even though my ovs configuration is fine. Refer to the output on the very bottom. Agents are running on both controller and compute node. Any help would be appreciated.

root@controller:~# tcpdump -i eth1 tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth1, link-type EN10MB (Ethernet), capture size 65535 bytes 12:00:55.884644 ARP, Request who-has 10.1.1.1 tell 10.1.1.2, length 46 12:00:56.880832 ARP, Request who-has 10.1.1.1 tell 10.1.1.2, length 46 12:00:57.880824 ARP, Request who-has 10.1.1.1 tell 10.1.1.2, length 46

[OVS] tenant_network_type = gre

network_vlan_ranges =

tunnel_id_ranges = 1:1000 integration_bridge = br-int tunnel_bridge = br-tun local_ip = 10.1.1.1 enable_tunneling = True [AGENT] root_helper = sudo /usr/bin/quantum-rootwrap /etc/quantum/rootwrap.conf

2012-12-01 22:53:04 -0600 answered a question Can't ping gateway of tenant router....

Thanks yong sheng gong, that solved my question.

2012-12-01 22:52:49 -0600 answered a question Can't ping gateway of tenant router....

Got it. Something was wrong with the ovs_quantum_plugin.ini file. There was a space before [OVS]. Thanks for that command.

2012-11-30 20:41:00 -0600 answered a question Can't ping gateway of tenant router....

Ok, this is something with the agent on the compute node. To install the openvswitch plugin on the compute node is this the only package I'm installing ---"apt-get install -y quantum-plugin-openvswitch-agent", because when I installed other quantum packages I noticed a python-pyparsing package was installed. Do I need that to successfully install and run the openvswitch agent? For some reason when I do a "ps -ef | grep quantum" on the compute node I'm not seeing it running.....

2012-11-28 01:34:27 -0600 answered a question Can't ping gateway of tenant router....

The interfaces on the left side are on the ComputeNode, and on the right side are the interfaces on the layer 3 agent.

2012-11-28 01:29:33 -0600 answered a question Can't ping gateway of tenant router....

Alright, that didn't work. But that's ok. Before I go adding things again. I have a question. If I were to ping the gateway that is on the layer 3 agent. What is the actual path of interfaces both physical and logical that data will go through? Is the br-tun in both bridges on the ComputeNode and on the node running the layer 3 agent? I tried adding it and it didn't work. Restarted agents and it still didn't work. If someone can amend the diagram I have below to reflect the correct flow of traffic I would appreciate it.

ComputeNode Node running Layer3 agent (vNic------->br-int------>eth1)-----GRE TUNNEL-----------(eth1-----br-int------qr-1549a07f-3a)

2012-11-26 05:08:19 -0600 answered a question Can't ping gateway of tenant router....

[OVS]

tenant_network_type = gre

network_vlan_ranges =

tunnel_id_ranges = 1:1000 integration_bridge = br-int tunnel_bridge = br-tun local_ip = 10.1.1.1 enable_tunneling = True

[AGENT]

root_helper = sudo /usr/bin/quantum-rootwrap /etc/quantum/rootwrap.conf

^Output from /etc/quantum/plugins/openvswitch/ovs_quantum_plugin.ini on (Controller)

Local IP was wrong, I changed it restarted services on both compute and controller, still not seeing it up. I'm at a loss. Getting close though.

2012-11-26 04:50:31 -0600 answered a question Can't ping gateway of tenant router....

Ya, I am using tunneling. You are right. I do not see a tunnel established. I'm at a loss. Not seeing anything useful in /var/log/quantum/openvswitch-agent.log on either the controller or compute node.

quantum 2842 1 1 14:55 ? 00:05:57 python /usr/bin/quantum-openvswitch-agent --config-file=/etc/quantum/quantum.conf --config-file=/etc/quantum/plugins/openvswitch/ovs_quantum_plugin.ini --log-file=/var/log/quantum/openvswitch-agent.log root 3852 6831 0 23:30 pts/2 00:00:00 grep --color=auto ovs

^Thats running on the controller

root@ComputeNode1:~# service quantum-plugin-openvswitch-agent start quantum-plugin-openvswitch-agent start/running, process 26738

^Compute Node


[OVS]

network_vlan_ranges =

tunnel_id_ranges = 1:1000 integration_bridge = br-int tunnel_bridge = br-tun local_ip = 10.1.1.2 enable_tunneling = True

[AGENT]

root_helper = sudo /usr/bin/quantum-rootwrap /etc/quantum/rootwrap.conf

^Output from /etc/quantum/plugins/openvswitch/ovs_quantum_plugin.ini on (Compute Node)

2012-11-26 04:02:06 -0600 asked a question Can't ping gateway of tenant router....

Just installed folsom on ubuntu 12.04, using this guide. http://docs.openstack.org/folsom/basic-install/content/basic-install_intro.html (http://docs.openstack.org/folsom/basi...)

I have one controller and one compute node. I used the script provided with the install and created a private network within a tenant. L3 agent is on the controller.

When I try to ping the gateway address of the private subnet, the arp gets to eth1 on the controller, and then gets to the bridge that eth1 is in, but the "qr-1549a07f-3a" interface never responds to the arp. This probably is something simple, but I am out of ideas for tonight. qr-1549a07f-3a is up Any help would be appreciated.

root@controller:~# ifconfig qr-1549a07f-3a qr-1549a07f-3a Link encap:Ethernet HWaddr 32:a4:d3:c9:72:e8
inet addr:10.5.5.1 Bcast:10.5.5.255 Mask:255.255.255.0 inet6 addr: fe80::30a4:d3ff:fec9:72e8/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 B) TX bytes:468 (468.0 B)

root@controller:~# ovs-vsctl show 195ff79e-e01a-4e93-84bc-868d8782f284 Bridge br-ex Port "qg-2b9a29bb-4b" Interface "qg-2b9a29bb-4b" type: internal Port "eth2" Interface "eth2" Port br-ex Interface br-ex type: internal Bridge br-int Port "tap4265fa27-08" tag: 1 Interface "tap4265fa27-08" type: internal Port "eth1" Interface "eth1" Port br-int Interface br-int type: internal Port "qr-1549a07f-3a" tag: 1 Interface "qr-1549a07f-3a" type: internal ovs_version: "1.4.0+build0"