Revision history [back]

click to hide/show revision 1
initial version

RDO MultiNode GRE tunnels & External network

Hello,

I'm trying to deply a multi-node openstack infrastructure with Neutron using GRE tunnels.

I have one controller node doing networking stuff and one QEMU compute node. I have three different networks for both host.

ETH0 is used for external access (Internet). Network 192.168.100.0/24, the gateway is 192.168.100.253.

ETH1 is used for management network (communication between Openstack components). Network 192.168.200.0/24

ETH2 is used for communication between VMs. Network 192.168.220.0/24

For now, communications between VMs on a private network works. However, the access on the external network fails. For example, I have a host with IP 192.168.100.14 trying to ping an openstack VM with a floating IP 192.168.100.171. The ping requests get no responses. I have made a tcpdump on both br-ex and eth0 of the controller/network node. I see no ICMP packet, just ARP requests on ETH0 and Br-ex, the ARP replies are only on ETH0...

My network interfaces on Controller/Network node are defined like this:

DEVICE=br-ex

DEVICETYPE=ovs

TYPE=OVSBridge

ONBOOT=yes

BOOTPROTO=static

IPADDR=192.168.100.101

NETMASK=255.255.255.0

GATEWAY=192.168.100.253

DEVICE=eth0

HWADDR=xxx

TYPE=OVSPort

DEVICETYPE=ovs

OVS_BRIDGE=br-ex

ONBOOT=yes

promisc=yes

DEVICE=eth2

HWADDR=xxx

TYPE=Ethernet

UUID=17998848-ca4f-4389-b1ab-3664dec55208

ONBOOT=yes

NM_CONTROLLED=no

BOOTPROTO=static

IPADDR=192.168.220.1

NETMASK=255.255.255.0

I don't know how to debug this network issue. Do you have an idea?

Thanks beforehand for your kind answers.

RDO MultiNode GRE tunnels & External network

Hello,

I'm trying to deply a multi-node openstack infrastructure with Neutron using GRE tunnels.

I have one controller node doing networking stuff and one QEMU compute node. I have three different networks for both host.

ETH0 is used for external access (Internet). Network 192.168.100.0/24, the gateway is 192.168.100.253.

ETH1 is used for management network (communication between Openstack components). Network 192.168.200.0/24

ETH2 is used for communication between VMs. Network 192.168.220.0/24

For now, communications between VMs on a private network works. However, the access on the external network fails. For example, I have a host with IP 192.168.100.14 trying to ping an openstack VM with a floating IP 192.168.100.171. The ping requests get no responses. I have made a tcpdump on both br-ex and eth0 of the controller/network node. I see no ICMP packet, just ARP requests on ETH0 and Br-ex, the ARP replies are only on ETH0...

My network interfaces on Controller/Network node are defined like this:

DEVICE=br-ex

DEVICETYPE=ovs

TYPE=OVSBridge

ONBOOT=yes

BOOTPROTO=static

IPADDR=192.168.100.101

NETMASK=255.255.255.0

GATEWAY=192.168.100.253

DEVICE=eth0

HWADDR=xxx

TYPE=OVSPort

DEVICETYPE=ovs

OVS_BRIDGE=br-ex

ONBOOT=yes

promisc=yes

DEVICE=eth2

HWADDR=xxx

TYPE=Ethernet

UUID=17998848-ca4f-4389-b1ab-3664dec55208

ONBOOT=yes

NM_CONTROLLED=no

BOOTPROTO=static

IPADDR=192.168.220.1

NETMASK=255.255.255.0

I don't know how to debug this network issue. Do you have an idea?

Thanks beforehand for your kind answers.

RDO MultiNode GRE tunnels & External network

Hello,

I'm trying to deply a multi-node openstack infrastructure with Neutron using GRE tunnels.

I have one controller node doing networking stuff and one QEMU compute node. I have three different networks for both host.

ETH0 is used for external access (Internet). Network 192.168.100.0/24, the gateway is 192.168.100.253.

ETH1 is used for management network (communication between Openstack components). Network 192.168.200.0/24

ETH2 is used for communication between VMs. Network 192.168.220.0/24

For now, communications between VMs on a private network works. However, the access on the external network fails. For example, I have a host with IP 192.168.100.14 trying to ping an openstack VM with a floating IP 192.168.100.171. The ping requests get no responses. I have made a tcpdump on both br-ex and eth0 of the controller/network node. I see no ICMP packet, just ARP requests on ETH0 and Br-ex, the ARP replies are only on ETH0...

My network interfaces on Controller/Network node are defined like this:

DEVICE=br-ex

DEVICETYPE=ovs

TYPE=OVSBridge

ONBOOT=yes

BOOTPROTO=static

IPADDR=192.168.100.101

NETMASK=255.255.255.0

GATEWAY=192.168.100.253

DEVICE=eth0

HWADDR=xxx

TYPE=OVSPort

DEVICETYPE=ovs

OVS_BRIDGE=br-ex

ONBOOT=yes

promisc=yes

DEVICE=eth2

HWADDR=xxx

TYPE=Ethernet

UUID=17998848-ca4f-4389-b1ab-3664dec55208

ONBOOT=yes

NM_CONTROLLED=no

BOOTPROTO=static

IPADDR=192.168.220.1

NETMASK=255.255.255.0

I don't know how to debug this network issue. Do you have an idea?

Thanks beforehand for your kind answers.