外部物理路由器可以通过配置bridge_mappings参数,另外想要在一个计算节点上同时使用flat, vlan, gre, vxlan网络需配置tenant_network_types = flat,vlan,gre,vxlan。
/etc/neutron/plugins/ml2/ml2_conf.ini
[ml2]
tenant_network_types = flat,vlan,gre,vxlan
type_drivers = local,flat,vlan,gre,vxlan
mechanism_drivers = openvswitch
[ovs]
bridge_mappings = physnet1:br-phy
使用devstack搭建环境的话,很简单,加上Q_ML2_TENANT_NETWORK_TYPE=flat,vlan,gre,vxlan 即可。
br-phy这个ovs网桥可以随便给它一个IP模拟,如172.16.1.1。当然,如果是真实物理网络的话,可以使用type=internal参数允许IP 172.16.1.1仍然设置在eth0上(别设置在br-phy上),这样不会对你的物理机的网络造成影响。
sudo ovs-vsctl -- --may-exist add-port br-phy eth0 -- set interface eth0 type=internal
demo@openstack:~$ grep -r '^enable_' /etc/neutron/dhcp_agent.ini enable_isolated_metadata = True enable_metadata_network = True demo@openstack:~$ grep -r '^enable_' /etc/neutron/l3_agent.ini enable_metadata_proxy = False
由于使用了外部物理网络,也就没有了l3-agent。由于还要使用neutron metadata,那么需要启动dhcp agent上的metadata proxy服务,但这要求使用隔离网络(最新的代码不用隔离子网也可以,但要指定force_metadata=True )。
所以隔离网络是指在neutron中有一个port连着subnet,且这个port的IP也是gatway的IP。所以在创建子网中指定--no-gateway方式创建的子网是隔离子网(neutron subnet-create isolate_net 10.0.3.0/24 --no-gateway --name=isolate_subnet),这种方式显然不符合使用外部网络的要求,下面虽然通过--gateway指定了网关,但由于是外部网关(--router:external=True),所以它实际也满足隔离子网的要求。
dhcp-agent上的metadata proxy与l3-agent上metadata proxy的最大不同是,dhcp-agent上的使用端口80,l3-agent上的使用端口9697,所以前者不需要到9697的DNAT规则。剩下的故事就和l3-agent上的metadata proxy一模一样了.
demo@openstack:~$ sudo ip netns exec qdhcp-ea600f11-4b53-4ab0-88e2-63f2fa2e2312 ip addr show tap16021052-cb
49: tap16021052-cb: <BROADCAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default
link/ether fa:16:3e:be:37:bb brd ff:ff:ff:ff:ff:ff
inet 172.16.1.102/24 brd 172.16.1.255 scope global tap16021052-cb
valid_lft forever preferred_lft forever
inet 169.254.169.254/16 brd 169.254.255.255 scope global tap16021052-cb
valid_lft forever preferred_lft forever
demo@openstack:~$ sudo ip netns exec qdhcp-ea600f11-4b53-4ab0-88e2-63f2fa2e2312 route -n |grep 169
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 tap16021052-cb
demo@openstack:~$ nova boot --flavor 1 --image cirros-0.3.4-x86_64-uec --nic net-id=ea600f11-4b53-4ab0-88e2-63f2fa2e2312 i1
demo@openstack:~/devstack$ nova list
+--------------------------------------+------+--------+------------+-------------+----------------------+
| ID | Name | Status | Task State | Power State | Networks |
+--------------------------------------+------+--------+------------+-------------+----------------------+
| fa0183ca-6369-4042-8f50-e9afea70df83 | i1 | ACTIVE | - | Running | phy_net=172.16.1.103 |
+--------------------------------------+------+--------+------------+-------------+----------------------
demo@openstack:~$ ping -c 1 172.16.1.103
PING 172.16.1.103 (172.16.1.103) 56(84) bytes of data.
64 bytes from 172.16.1.103: icmp_seq=1 ttl=64 time=0.474 ms
--- 172.16.1.103 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.474/0.474/0.474/0.000 ms
demo@openstack:~$ ssh [email protected]
The authenticity of host '172.16.1.103 (172.16.1.103)' can't be established.
RSA key fingerprint is b3:49:13:b3:a0:13:de:a6:6e:18:14:cb:29:8a:8a:db.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '172.16.1.103' (RSA) to the list of known hosts.
[email protected]'s password:
$ route -n |grep 169
169.254.169.254 172.16.1.102 255.255.255.255 UGH 0 0 0 eth0
$ wget http://169.254.169.254/openstack
Connecting to 169.254.169.254 (169.254.169.254:80)
openstack 100% |****************************************************| 39 0:00:00 ETA
$
demo@openstack:~$ sudo cat /opt/stack/data/neutron/dhcp/ea600f11-4b53-4ab0-88e2-63f2fa2e2312/opts |grep 169
tag:tag0,option:classless-static-route,169.254.169.254/32,172.16.1.102,0.0.0.0/0,172.16.1.1
tag:tag0,249,169.254.169.254/32,172.16.1.102,0.0.0.0/0,172.16.1.1
用外部物理路由器时与外部dhcp服务时如何使用metadata服务(by quqi99)
http://blog.csdn.net/quqi99/article/details/50440347