Skip to content

Latest commit

 

History

History
323 lines (282 loc) · 12.9 KB

README.md

File metadata and controls

323 lines (282 loc) · 12.9 KB
Raw

OpenVNet test tools with network namespace

Tested Environment

  • CentOS 6.5
  • kernel-2.6.32-431.1.2.0.1.el6.x86_64 (CentOS 6.5 latest)

Network Namespace

In CentOS 6.x, network namespace is not supported, so we need to additional packages to use network namespace. OpenStack RDO provides network namespace aware kernel and iproute.

CentOS 6.5

In CentOS 6.5, kernel itself supports network namespace, but iproute does not support network namespace (no "netns" subcommand).

The following commands setup OpenStack RDO release repository to /etc/yum.repos.d and "yum update" installs a new version of iproute (which supports network namespace),

# yum install http://rdo.fedorapeople.org/rdo-release.rpm
# yum install iproute

or you can install iproute package explicitly.

# yum install http://rdo.fedorapeople.org/rdo-release.rpm

After the installation, check the appropriate iproute package is installed and it works.

# rpm -qa | grep iproute
iproute-2.6.32-130.el6ost.netns.2.x86_64
# ip netns
# ip netns add testns1
# ip netns
testns1
# ip netns delete testns1

CentOS 6.4 or older

I have not tested these version, but ROD provides network namespace aware kernel. The packages are available at: http://repos.fedorapeople.org/repos/openstack/openstack-havana/epel-6/

How to run

Install OpenVNet

Follow OpenVNet install guide: https://github.com/axsh/openvnet/blob/master/InstallGuide.md No need to go "Let's try 1Box OpenVNet" section.

Make sure to start redis before starting OpenVNet services:

# chkconfig redis on
# service redis start

Try OpenVNet with network namespaces

Register various OpenVNet resources.

# ./vnet-register-all.sh
  • vnet-register-datapath.sh : Create OpenFlow datapath
  • vnet-register-net1.sh : Create three virtual networks net1
    • net1 : 172.16.1.0/24
  • vnet-register-net2.sh : Create three virtual networks net2
    • net2 : 172.16.2.0/24
  • vnet-register-net3.sh : Create three virtual networks net3
    • net3 : 172.16.1.0/24 (IP address range is overlaped with net1)
  • vnet-register-router.sh
    • Create a router between net1 and net2

Setup OVS datapath and create interfaces connected to OVS bridge

# ./ovs-create.sh
# ./if-create.sh

After running the above scripts, you can see the following. You can see 6 network namespaces and 6 veth interfaces are connected to OVS bridge br0. OVS bridge br0 is connected to OpenVNet agent (vnet-vna) with OpenFlow protocol (is_conncted: true).

# ip netns
ns5
ns2
ns6
ns4
ns3
ns1
# ip -o link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN \    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 52:54:00:fd:9f:45 brd ff:ff:ff:ff:ff:ff
4: ovs-system: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN \    link/ether 4e:5e:dd:a0:1a:fe brd ff:ff:ff:ff:ff:ff
124: br0: <BROADCAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN \    link/ether 02:fd:ee:8e:0a:44 brd ff:ff:ff:ff:ff:ff
127: if-veth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 1a:42:f4:d5:d4:77 brd ff:ff:ff:ff:ff:ff
130: if-veth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether b2:d3:e4:dc:df:f1 brd ff:ff:ff:ff:ff:ff
133: if-veth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether c6:3f:5d:4b:b1:38 brd ff:ff:ff:ff:ff:ff
136: if-veth4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether ce:50:9d:28:81:81 brd ff:ff:ff:ff:ff:ff
139: if-veth5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 2a:c0:e2:e8:29:94 brd ff:ff:ff:ff:ff:ff
142: if-veth6: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether b2:9c:63:8a:d8:c8 brd ff:ff:ff:ff:ff:ff
# ovs-vsctl show
73ab1cf2-5964-40d4-9dfe-eb96407fafd2
    Bridge "br0"
        Controller "tcp:127.0.0.1:6633"
            is_connected: true
        fail_mode: standalone
        Port "if-veth4"
            Interface "if-veth4"
        Port "if-veth6"
            Interface "if-veth6"
        Port "if-veth1"
            Interface "if-veth1"
        Port "if-veth2"
            Interface "if-veth2"
        Port "if-veth5"
            Interface "if-veth5"
        Port "br0"
            Interface "br0"
                type: internal
        Port "if-veth3"
            Interface "if-veth3"
    ovs_version: "1.11.0"

Check network connectivity

  • net1
# ip netns exec ns1 ip -o addr
125: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN \    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
125: lo    inet 127.0.0.1/8 scope host lo
125: lo    inet6 ::1/128 scope host \       valid_lft forever preferred_lft forever
126: ns-veth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 52:54:00:0d:84:01 brd ff:ff:ff:ff:ff:ff
126: ns-veth1    inet 172.16.1.1/24 brd 172.16.1.255 scope global ns-veth1
126: ns-veth1    inet6 fe80::5054:ff:fe0d:8401/64 scope link \       valid_lft forever preferred_lft forever
# ip netns exec ns2 ip -o addr
128: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN \    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
128: lo    inet 127.0.0.1/8 scope host lo
128: lo    inet6 ::1/128 scope host \       valid_lft forever preferred_lft forever
129: ns-veth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 52:54:00:0d:84:02 brd ff:ff:ff:ff:ff:ff
129: ns-veth2    inet 172.16.1.2/24 brd 172.16.1.255 scope global ns-veth2
129: ns-veth2    inet6 fe80::5054:ff:fe0d:8402/64 scope link \       valid_lft forever preferred_lft forever
# ip netns exec ns1 arp -na
# ip netns exec ns1 ping 172.16.1.2
PING 172.16.1.2 (172.16.1.2) 56(84) bytes of data.
64 bytes from 172.16.1.2: icmp_seq=1 ttl=64 time=0.391 ms
64 bytes from 172.16.1.2: icmp_seq=2 ttl=64 time=0.049 ms
64 bytes from 172.16.1.2: icmp_seq=3 ttl=64 time=0.044 ms
^C
--- 172.16.1.2 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2441ms
rtt min/avg/max/mdev = 0.044/0.161/0.391/0.162 ms
# ip netns exec ns1 arp -na
? (172.16.1.2) at 52:54:00:0d:84:02 [ether] on ns-veth1
# ip netns exec ns2 arp -na
? (172.16.1.1) at 52:54:00:0d:84:01 [ether] on ns-veth2
  • net2
# ip netns exec ns3 ip -o addr
131: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN \    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
131: lo    inet 127.0.0.1/8 scope host lo
131: lo    inet6 ::1/128 scope host \       valid_lft forever preferred_lft forever
132: ns-veth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 52:54:00:0d:84:03 brd ff:ff:ff:ff:ff:ff
132: ns-veth3    inet 172.16.2.3/24 brd 172.16.2.255 scope global ns-veth3
132: ns-veth3    inet6 fe80::5054:ff:fe0d:8403/64 scope link \       valid_lft forever preferred_lft forever
# ip netns exec ns4 ip -o addr
134: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN \    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
134: lo    inet 127.0.0.1/8 scope host lo
134: lo    inet6 ::1/128 scope host \       valid_lft forever preferred_lft forever
135: ns-veth4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000\    link/ether 52:54:00:0d:84:04 brd ff:ff:ff:ff:ff:ff
135: ns-veth4    inet 172.16.2.4/24 brd 172.16.2.255 scope global ns-veth4
135: ns-veth4    inet6 fe80::5054:ff:fe0d:8404/64 scope link \       valid_lft forever preferred_lft forever
# ip netns exec ns3 ping 172.16.2.4
PING 172.16.2.4 (172.16.2.4) 56(84) bytes of data.
64 bytes from 172.16.2.4: icmp_seq=1 ttl=64 time=0.713 ms
64 bytes from 172.16.2.4: icmp_seq=2 ttl=64 time=0.086 ms
64 bytes from 172.16.2.4: icmp_seq=3 ttl=64 time=0.058 ms
64 bytes from 172.16.2.4: icmp_seq=4 ttl=64 time=0.041 ms
64 bytes from 172.16.2.4: icmp_seq=5 ttl=64 time=0.045 ms
^[^C
--- 172.16.2.4 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 4219ms
rtt min/avg/max/mdev = 0.041/0.188/0.713/0.263 ms
  • net3

Before ping from 172.16.1.1 to 172.16.1.2, start tcpdump on interfaces both on net1 and net3.

# ip netns exec ns2 tcpdump -i ns-veth2

# ip netns exec ns6 tcpdump -i ns-veth6

Ping from 172.16.1.1 (ns5) to 172.16.1.2 (ns6)

# ip netns exec ns5 ping -c 5 172.16.1.2
PING 172.16.1.2 (172.16.1.2) 56(84) bytes of data.
64 bytes from 172.16.1.2: icmp_seq=1 ttl=64 time=0.391 ms
64 bytes from 172.16.1.2: icmp_seq=2 ttl=64 time=0.052 ms
64 bytes from 172.16.1.2: icmp_seq=3 ttl=64 time=0.087 ms
64 bytes from 172.16.1.2: icmp_seq=4 ttl=64 time=0.048 ms
64 bytes from 172.16.1.2: icmp_seq=5 ttl=64 time=0.045 ms

--- 172.16.1.2 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 4000ms
rtt min/avg/max/mdev = 0.045/0.124/0.391/0.134 ms

We can see packets in tcpdump on ns6.

# ip netns exec ns6 tcpdump -i ns-veth6
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on ns-veth6, link-type EN10MB (Ethernet), capture size 65535 bytes
21:57:40.246563 ARP, Request who-has 172.16.1.2 tell 172.16.1.1, length 28
21:57:40.246581 ARP, Reply 172.16.1.2 is-at 52:54:00:0d:84:06 (oui Unknown), length 28
21:57:40.246646 IP 172.16.1.1 > 172.16.1.2: ICMP echo request, id 22894, seq 1, length 64
21:57:40.246658 IP 172.16.1.2 > 172.16.1.1: ICMP echo reply, id 22894, seq 1, length 64
21:57:41.246538 IP 172.16.1.1 > 172.16.1.2: ICMP echo request, id 22894, seq 2, length 64
21:57:41.246556 IP 172.16.1.2 > 172.16.1.1: ICMP echo reply, id 22894, seq 2, length 64
21:57:42.246522 IP 172.16.1.1 > 172.16.1.2: ICMP echo request, id 22894, seq 3, length 64
21:57:42.246572 IP 172.16.1.2 > 172.16.1.1: ICMP echo reply, id 22894, seq 3, length 64
21:57:43.246516 IP 172.16.1.1 > 172.16.1.2: ICMP echo request, id 22894, seq 4, length 64
21:57:43.246532 IP 172.16.1.2 > 172.16.1.1: ICMP echo reply, id 22894, seq 4, length 64
21:57:44.246490 IP 172.16.1.1 > 172.16.1.2: ICMP echo request, id 22894, seq 5, length 64
21:57:44.246505 IP 172.16.1.2 > 172.16.1.1: ICMP echo reply, id 22894, seq 5, length 64
21:57:45.246479 ARP, Request who-has 172.16.1.1 tell 172.16.1.2, length 28
21:57:45.246678 ARP, Reply 172.16.1.1 is-at 52:54:00:0d:84:05 (oui Unknown), length 28
^C
14 packets captured
14 packets received by filter
0 packets dropped by kernel

Routing between net1 and net2

We need to setup additional route information in each network namespaces. It is already done by if-create.sh script.

ping from 172.16.1.1 (ns1) to 172.16.2.3 (ns3). You will see ARP replies from 172.16.2.3 if it works well.

# ip netns exec ns1 netstat -nr
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
172.16.2.0      172.16.1.254    255.255.255.0   UG        0 0          0 ns-veth1
172.16.1.0      0.0.0.0         255.255.255.0   U         0 0          0 ns-veth1
# ip netns exec ns3 netstat -nr
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
172.16.2.0      0.0.0.0         255.255.255.0   U         0 0          0 ns-veth3
172.16.1.0      172.16.2.254    255.255.255.0   UG        0 0          0 ns-veth3
# ip netns exec ns1 ping 172.16.2.3
PING 172.16.2.3 (172.16.2.3) 56(84) bytes of data.
64 bytes from 172.16.2.3: icmp_seq=1 ttl=64 time=18.1 ms
64 bytes from 172.16.2.3: icmp_seq=2 ttl=64 time=0.056 ms
64 bytes from 172.16.2.3: icmp_seq=3 ttl=64 time=0.048 ms
^C
--- 172.16.2.3 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2790ms
rtt min/avg/max/mdev = 0.048/6.089/18.163/8.537 ms
# ip netns exec ns1 arp -na
? (172.16.1.254) at 52:54:00:74:00:00 [ether] on ns-veth1
# ip netns exec ns3 arp -na
? (172.16.2.254) at 52:54:00:74:22:22 [ether] on ns-veth3

Utilities

Reinitialize OpenVNet and network namespaces.

# ./if-delete.sh
# ./ovs-delete.sh
# ./vnet-all.sh resetdb

"vnet-all.sh resetdb" does the following:

  • Stop all openvnet services
  • Drop "vnet" database
  • Create "vnet" database
  • Populate openvnet tables to "vnet" database
  • Remove OpenVNet logs in /var/log/wakame-vnet
  • Start all openvnet services

vnctl.sh

It is a wrapper of /opt/axsh/wakame-vnet/vnctl/bin/vnctl. After running vnctl command, it formats the outputs of vnctl to human-friendly format. If you want raw output of vnctl, please comment out DO_FORMAT=True.

# ./vnctl.sh datapath show
{"id"=>1,
 "uuid"=>"dp-br0",
 "display_name"=>"br0",
 "is_connected"=>false,
 "dpid"=>"0x000002fdee8e0a44",
 "dc_segment_id"=>nil,
 "ipv4_address"=>3232266959,
 "node_id"=>"vnet3",
 "created_at"=>"2013-12-23T13:08:11Z",
 "updated_at"=>"2013-12-23T13:08:11Z",
 "dc_segment"=>nil}

dump-flows.sh

It is a wrapper of /opt/axsh/wakame-vnet/vnctl/bin/vnflows. OpenVNet installs MANY MANY flow entries to OVS bridge :-) and it helps us to debug flows.

It just runns the following:

# ovs-ofctl dump-flows br0 | ./bin/vnflows