在前面的文章中，已经完成了devstack的安装。下面，我会介绍如何使用neutron
- 首先创建两个neutron net, vmnet1 和vmnet2
stack@ubuntu:~/devstack$ source openrc admin admin stack@ubuntu:~/devstack$ neutron net-list +--------------------------------------+---------+-------------------------------------------------------+ | id | name | subnets | +--------------------------------------+---------+-------------------------------------------------------+ | 5365b56f-9544-40d4-b065-a9d50126b55b | private | fc3a839e-412c-4f0d-b1ed-87be1a7b0bf8 192.168.106.0/24 | | a101e11f-7bbe-4f4d-a52e-5bb312d5f8b9 | public | 570e8b64-8332-41f7-8780-12d9bc54f195 172.24.4.0/24 | +--------------------------------------+---------+-------------------------------------------------------+ stack@ubuntu:~/devstack$ neutron net-create vmnet1 Created a new network: +---------------------------+--------------------------------------+ | Field | Value | +---------------------------+--------------------------------------+ | admin_state_up | True | | id | 54028daa-9158-4fc7-9f97-d400dc801e97 | | name | vmnet1 | | provider:network_type | vlan | | provider:physical_network | physnet1 | | provider:segmentation_id | 352 | | router:external | False | | shared | False | | status | ACTIVE | | subnets | | | tenant_id | dcf6685a3e304e4c92af08bca762aa33 | +---------------------------+--------------------------------------+ stack@ubuntu:~/devstack$ neutron net-create vmnet2 Created a new network: +---------------------------+--------------------------------------+ | Field | Value | +---------------------------+--------------------------------------+ | admin_state_up | True | | id | 7b0394ec-34c0-48d3-a17d-82ef9bbc2cd9 | | name | vmnet2 | | provider:network_type | vlan | | provider:physical_network | physnet1 | | provider:segmentation_id | 353 | | router:external | False | | shared | False | | status | ACTIVE | | subnets | | | tenant_id | dcf6685a3e304e4c92af08bca762aa33 | +---------------------------+--------------------------------------+ stack@ubuntu:~/devstack$ neutron net-list +--------------------------------------+---------+-------------------------------------------------------+ | id | name | subnets | +--------------------------------------+---------+-------------------------------------------------------+ | 5365b56f-9544-40d4-b065-a9d50126b55b | private | fc3a839e-412c-4f0d-b1ed-87be1a7b0bf8 192.168.106.0/24 | | a101e11f-7bbe-4f4d-a52e-5bb312d5f8b9 | public | 570e8b64-8332-41f7-8780-12d9bc54f195 172.24.4.0/24 | | 54028daa-9158-4fc7-9f97-d400dc801e97 | vmnet1 | | | 7b0394ec-34c0-48d3-a17d-82ef9bbc2cd9 | vmnet2 | | +--------------------------------------+---------+-------------------------------------------------------+
- 现在两个网络就创建好了，但现在其实这是两个独立的网络，是不互通的，接下来创建router： router-vmnet1-vmnet2
stack@ubuntu:~/devstack$ neutron router-create router-vmnet1-vmnet2 Created a new router: +-----------------------+--------------------------------------+ | Field | Value | +-----------------------+--------------------------------------+ | admin_state_up | True | | distributed | False | | external_gateway_info | | | ha | False | | id | 4bef86df-3b13-4b23-80f0-3a2bd8d0c746 | | name | router-vmnet1-vmnet2 | | routes | | | status | ACTIVE | | tenant_id | dcf6685a3e304e4c92af08bca762aa33 | +-----------------------+--------------------------------------+
- 在vmnet1和vmnet2中，分别创建subnet
stack@ubuntu:~/devstack$ neutron subnet-create 54028daa-9158-4fc7-9f97-d400dc801e97 30.30.30.0/24 Created a new subnet: +-------------------+------------------------------------------------+ | Field | Value | +-------------------+------------------------------------------------+ | allocation_pools | {"start": "30.30.30.2", "end": "30.30.30.254"} | | cidr | 30.30.30.0/24 | | dns_nameservers | | | enable_dhcp | True | | gateway_ip | 30.30.30.1 | | host_routes | | | id | c0be451f-91a1-4f9f-a29b-13adea79798e | | ip_version | 4 | | ipv6_address_mode | | | ipv6_ra_mode | | | name | | | network_id | 54028daa-9158-4fc7-9f97-d400dc801e97 | | tenant_id | dcf6685a3e304e4c92af08bca762aa33 | +-------------------+------------------------------------------------+ stack@ubuntu:~/devstack$ neutron subnet-create 7b0394ec-34c0-48d3-a17d-82ef9bbc2cd9 40.40.40.0/24 Created a new subnet: +-------------------+------------------------------------------------+ | Field | Value | +-------------------+------------------------------------------------+ | allocation_pools | {"start": "40.40.40.2", "end": "40.40.40.254"} | | cidr | 40.40.40.0/24 | | dns_nameservers | | | enable_dhcp | True | | gateway_ip | 40.40.40.1 | | host_routes | | | id | 0b9c9ad8-3309-4c96-afda-ecb7e53dc519 | | ip_version | 4 | | ipv6_address_mode | | | ipv6_ra_mode | | | name | | | network_id | 7b0394ec-34c0-48d3-a17d-82ef9bbc2cd9 | | tenant_id | dcf6685a3e304e4c92af08bca762aa33 | +-------------------+------------------------------------------------+
- 接下来，给router加入interface，分别连接到vmnet1的subnet和vmnet2的subnet
stack@ubuntu:~/devstack$ neutron router-interface-add 4bef86df-3b13-4b23-80f0-3a2bd8d0c746 subnet=c0be451f-91a1-4f9f-a29b-13adea79798e Added interface 8fc6f138-150c-47de-8c88-fa924ce2302d to router 4bef86df-3b13-4b23-80f0-3a2bd8d0c746. stack@ubuntu:~/devstack$ neutron router-interface-add 4bef86df-3b13-4b23-80f0-3a2bd8d0c746 subnet=0b9c9ad8-3309-4c96-afda-ecb7e53dc519 Added interface 4e830e98-e24b-45dd-bc6e-338e47b31b1b to router 4bef86df-3b13-4b23-80f0-3a2bd8d0c746.
- 然后，我们看，Linux 网络配置上的变化，首先是network namespace， 你会发现名为qrouter-4bef86df-3b13-4b23-80f0-3a2bd8d0c746 的Linux Namespace创建成功
stack@ubuntu:~/devstack$ sudo ip netns exec qrouter-4bef86df-3b13-4b23-80f0-3a2bd8d0c746 ip addr 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 15: qr-4e830e98-e2: <BROADCAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default link/ether fa:16:3e:8c:4d:c7 brd ff:ff:ff:ff:ff:ff inet 40.40.40.1/24 brd 40.40.40.255 scope global qr-4e830e98-e2 valid_lft forever preferred_lft forever inet6 fe80::f816:3eff:fe8c:4dc7/64 scope link valid_lft forever preferred_lft forever 16: qr-8fc6f138-15: <BROADCAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default link/ether fa:16:3e:67:b5:14 brd ff:ff:ff:ff:ff:ff inet 30.30.30.1/24 brd 30.30.30.255 scope global qr-8fc6f138-15 valid_lft forever preferred_lft forever inet6 fe80::f816:3eff:fe67:b514/64 scope link valid_lft forever preferred_lft forever
- 上图的qr-4e830e98-e2和qr-8fc6f138-15，可以理解为vmnet1和vmnet2的gateway， 同时它们也都连接在ovs的br-int bridge上，如下：
stack@ubuntu:~/devstack$ sudo ovs-vsctl show ....... Bridge br-int fail_mode: secure Port patch-tun Interface patch-tun type: patch options: {peer=patch-int} Port "qr-4e830e98-e2" tag: 3 Interface "qr-4e830e98-e2" type: internal Port br-int Interface br-int type: internal Port "int-br-eth1" Interface "int-br-eth1" type: patch options: {peer="phy-br-eth1"} Port "qr-8fc6f138-15" tag: 2 Interface "qr-8fc6f138-15" type: internal Port "qr-a6466103-17" tag: 1 Interface "qr-a6466103-17" type: internal
从上面的输出可以看出"qr-4e830e98-e2" is tagged with 3, "qr-8fc6f138-15"is tagged with 2
- 接下来，我要在vmnet1和vmnet2上分别创建一个virtual machine instance，看看openstack neutron中发生了什么！
#注意，这个一定要先增加这个，不然，后面的vm启动无法得到IP， 参见我的异常处理部分 stack@ubuntu:~/devstack$ nova secgroup-add-rule default icmp -1 -1 0.0.0.0/0 stack@ubuntu:~/devstack$ nova secgroup-add-rule default tcp 22 22 0.0.0.0/0
#创建两个vm
stack@ubuntu:~/devstack$ nova boot --image 54bdfba1-27f7-4fae-99f1-cfbbca78c9bd --flavor m1.tiny --nic net-id=54028daa-9158-4fc7-9f97-d400dc801e97 vm1
stack@ubuntu:~/devstack$ nova boot --image 54bdfba1-27f7-4fae-99f1-cfbbca78c9bd --flavor m1.tiny --nic net-id=7b0394ec-34c0-48d3-a17d-82ef9bbc2cd9 vm2
#查看vm是否ready
stack@ubuntu:~/devstack$ nova list
+--------------------------------------+------+--------+------------+-------------+-------------------+
| ID | Name | Status | Task State | Power State | Networks |
+--------------------------------------+------+--------+------------+-------------+-------------------+
| b22e706b-a1aa-4f73-9bd5-5f70bbf7e299 | vm1 | ACTIVE | - | Running | vmnet1=30.30.30.2 |
| 1f4fad7d-8eb9-4011-b196-337dea3d7d66 | vm2 | ACTIVE | - | Running | vmnet2=40.40.40.2 |
+--------------------------------------+------+--------+------------+-------------+-------------------+
现在这两个ip 30.30.30.2 和40.40.40.2是无法直接ping同的，因为是被Linux Namespace隔离的
stack@ubuntu:~/devstack$ ping 30.30.30.2 PING 30.30.30.2 (30.30.30.2) 56(84) bytes of data. ^C --- 30.30.30.2 ping statistics --- 4 packets transmitted, 0 received, 100% packet loss, time 3000ms stack@ubuntu:~/devstack$ ping 40.40.40.2 PING 40.40.40.2 (40.40.40.2) 56(84) bytes of data. ^C --- 40.40.40.2 ping statistics --- 4 packets transmitted, 0 received, 100% packet loss, time 3000ms
- 你可以这样在主机上ssh到vm上，这里我们以ssh到vm1为例：首先确定vm1所在的network为vmnet1， vmnet1对应的namespace为qdhcp-54028daa-9158-4fc7-9f97-d400dc801e97，然后
ip netns exec qqdhcp-54028daa-9158-4fc7-9f97-d400dc801e97 ssh cirros@30.30.30.2
密码为 cubswin:)
这样一个基于neutron network的devstack环境就搭建好了，大家有兴趣的话，可以一起探讨关于neutron跟多的东西，推荐大家几篇oracle上的文章，讲的深入浅出的，很不错
https://blogs.oracle.com/ronen/entry/diving_into_openstack_network_architecture
https://blogs.oracle.com/ronen/entry/diving_into_openstack_network_architecture1
https://blogs.oracle.com/ronen/entry/diving_into_openstack_network_architecture2
https://blogs.oracle.com/ronen/entry/diving_into_openstack_network_architecture3
https://blogs.oracle.com/ronen/entry/running_openstack_icehouse_with_zfs