Icehouse neutron can't access to instance with floating IP
Hi, All. I have deployed Openstack icehouse version on RHEL 6.5 for test lab followed by official guideline. After neutron configuration and initiate VM, I can't access to floating IP of VM. Nova instance have two interface for public and private network, I can't access neither of them. So, I can't reach instance. I have tried restart all things and re-configure neutron network. I'm not really familiar with openvswitch and networking, I have thought that i'm missing some important thing. Please help me.
I have two node configuration, first node act as controller, network node and compute node. and second node act as compute node.
Each node have two network interface, one for external and other for private.
**Controller Node and network node and compute node 1 (with External network IP -10.183.120.170)
[root@osnode01 neutron]# ifconfig -a
br-ex Link encap:Ethernet HWaddr 68:B5:99:C7:D2:8C
inet addr:10.183.120.170 Bcast:10.183.120.191 Mask:255.255.255.192
inet6 addr: fe80::fc72:2dff:fe52:4709/64 Scope:Link
UP BROADCAST RUNNING MTU:1500 Metric:1
RX packets:4698305 errors:0 dropped:0 overruns:0 frame:0
TX packets:4369482 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:1803665841 (1.6 GiB) TX bytes:1833241255 (1.7 GiB)
br-int Link encap:Ethernet HWaddr 8E:B9:63:DC:EE:45
inet6 addr: fe80::2c21:cff:fe90:d4fa/64 Scope:Link
UP BROADCAST RUNNING MTU:1500 Metric:1
RX packets:536 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:32184 (31.4 KiB) TX bytes:468 (468.0 b)
br-tun Link encap:Ethernet HWaddr D6:F1:17:0D:CA:4E
inet6 addr: fe80::8024:fdff:fe14:af81/64 Scope:Link
UP BROADCAST RUNNING MTU:1500 Metric:1
RX packets:230 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:15164 (14.8 KiB) TX bytes:468 (468.0 b)
eth0 Link encap:Ethernet HWaddr 68:B5:99:C7:D2:8C
inet6 addr: fe80::6ab5:99ff:fec7:d28c/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:5101773 errors:0 dropped:0 overruns:0 frame:0
TX packets:4779571 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1851895406 (1.7 GiB) TX bytes:1879660396 (1.7 GiB)
eth1 Link encap:Ethernet HWaddr 68:B5:99:C7:D2:8E
inet addr:192.168.255.10 Bcast:192.168.255.255 Mask:255.255.255.0
inet6 addr: fe80::6ab5:99ff:fec7:d28e/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:792 errors:0 dropped:0 overruns:0 frame:0
TX packets:812 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000 ...
Please,
1. post
ovs-vsctl show
andifconfig
- on both nodes.On Controller :-
2. neutron net-list
3. neutron router-list
4. ip netns
I have updated content what you mentioned. ovs qr and qg interface LINK_DOWN status is normal behavior because of netns? I've tried to manually create namespace and ovs interface, but link may also change to down status right after move ovs interface to spcific namespace.
Please, view link Router Interfaces Down . Your issues seem similar to described there.
You also wrote : I've tried to manually create namespace and ovs interface, but link may also change to down status right after move ovs interface to spcific namespace. You are not supposed to do that.