On Tue, Nov 5, 2019 at 6:14 PM shuangyang qian <[email protected]> wrote:
>
>
> cc to ovs-discuss
> ---------- Forwarded message ---------
> 发件人： shuangyang qian <[email protected]>
> Date: 2019年11月5日周二 下午6:12
> Subject: Re: [ovs-discuss] the network performence is not normal when use
> openvswitch.ko make from ovs tree
> To: Tonghao Zhang <[email protected]>
>
>
> thank you for your reply, i just change my kernel version as same as you and
> do the steps you provide, and get the same result which i metioned at first.
> The process is like below.
> on node1:
> # ovs-vsctl show
> 4f4b936e-ddb9-4fc6-b0aa-6eb6034d4671
> Bridge br-int
> Port br-int
> Interface br-int
> type: internal
> Port "gnv0"
> Interface "gnv0"
> type: geneve
> options: {csum="true", key="100", remote_ip="10.18.124.2"}
> Port "veth-vm1"
> Interface "veth-vm1"
> ovs_version: "2.12.0"
> # ip netns exec vm1 ip a
> 1: lo: <LOOPBACK> mtu 65536 qdisc noop state DOWN group default qlen 1000
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> 2: ovs-gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN
> group default qlen 1000
> link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
Why this netdev in your netns? you run the ovs in your netns ? the ovs
should running on host.
> 3: erspan0@NONE: <BROADCAST,MULTICAST> mtu 1450 qdisc noop state DOWN group
> default qlen 1000
> link/ether 32:d9:4f:86:c3:58 brd ff:ff:ff:ff:ff:ff
?
> 4: ovs-ip6gre0@NONE: <NOARP> mtu 1448 qdisc noop state DOWN group default
> qlen 1000
> link/gre6 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00 brd
> 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00
> 5: ovs-ip6tnl0@NONE: <NOARP> mtu 1452 qdisc noop state DOWN group default
> qlen 1000
?
> link/tunnel6 :: brd ::
> 19: vm1-eth0@if18: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue
> state UP group default qlen 1000
> link/ether 32:4b:51:e2:2b:f4 brd ff:ff:ff:ff:ff:ff link-netnsid 0
> inet 192.168.100.10/24 scope global vm1-eth0
> valid_lft forever preferred_lft forever
> inet6 fe80::304b:51ff:fee2:2bf4/64 scope link
> valid_lft forever preferred_lft forever
please set vm1-eth0 mtu to 1450.
> on node2:
> # ovs-vsctl show
> 53df6c21-c210-4c2c-a7ab-b1edb0df4a31
> Bridge br-int
> Port "veth-vm2"
> Interface "veth-vm2"
> Port "gnv0"
> Interface "gnv0"
> type: geneve
> options: {csum="true", key="100", remote_ip="10.18.124.1"}
> Port br-int
> Interface br-int
> type: internal
> ovs_version: "2.12.0"
> # ip netns exec vm2 ip a
> 1: lo: <LOOPBACK> mtu 65536 qdisc noop state DOWN group default qlen 1000
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> 2: ovs-gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN
> group default qlen 1000
> link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
> 3: erspan0@NONE: <BROADCAST,MULTICAST> mtu 1450 qdisc noop state DOWN group
> default qlen 1000
> link/ether 8e:90:3e:95:1b:dd brd ff:ff:ff:ff:ff:ff
> 4: ovs-ip6gre0@NONE: <NOARP> mtu 1448 qdisc noop state DOWN group default
> qlen 1000
> link/gre6 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00 brd
> 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00
> 5: ovs-ip6tnl0@NONE: <NOARP> mtu 1452 qdisc noop state DOWN group default
> qlen 1000
> link/tunnel6 :: brd ::
> 11: vm2-eth0@if10: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue
> state UP group default qlen 1000
> link/ether ee:e4:3e:16:6f:66 brd ff:ff:ff:ff:ff:ff link-netnsid 0
> inet 192.168.100.20/24 scope global vm2-eth0
> valid_lft forever preferred_lft forever
> inet6 fe80::ece4:3eff:fe16:6f66/64 scope link
> valid_lft forever preferred_lft forever
>
> and in network namespace vm1 on node1 i start iperf3 as server:
> # ip netns exec vm1 iperf3 -s
>
> and in network namespace vm2 on noed2 i start iperf3 as client:
> # ip netns exec vm2 iperf3 -c 192.168.100.10 -i 2 -t 10
> Connecting to host 192.168.100.10, port 5201
> [ 4] local 192.168.100.20 port 35258 connected to 192.168.100.10 port 5201
> [ ID] Interval Transfer Bandwidth Retr Cwnd
> [ 4] 0.00-2.00 sec 494 MBytes 2.07 Gbits/sec 151 952 KBytes
> [ 4] 2.00-4.00 sec 582 MBytes 2.44 Gbits/sec 3 1007 KBytes
> [ 4] 4.00-6.00 sec 639 MBytes 2.68 Gbits/sec 0 1.36 MBytes
> [ 4] 6.00-8.00 sec 618 MBytes 2.59 Gbits/sec 0 1.64 MBytes
> [ 4] 8.00-10.00 sec 614 MBytes 2.57 Gbits/sec 0 1.88 MBytes
> - - - - - - - - - - - - - - - - - - - - - - - - -
> [ ID] Interval Transfer Bandwidth Retr
> [ 4] 0.00-10.00 sec 2.88 GBytes 2.47 Gbits/sec 154 sender
> [ 4] 0.00-10.00 sec 2.88 GBytes 2.47 Gbits/sec receiver
>
> iperf Done.
>
> the openvswitch.ko in both two nodes is:
> # modinfo openvswitch
> filename:
> /lib/modules/3.10.0-957.el7.x86_64/extra/openvswitch/openvswitch.ko
> alias: net-pf-16-proto-16-family-ovs_ct_limit
> alias: net-pf-16-proto-16-family-ovs_meter
> alias: net-pf-16-proto-16-family-ovs_packet
> alias: net-pf-16-proto-16-family-ovs_flow
> alias: net-pf-16-proto-16-family-ovs_vport
> alias: net-pf-16-proto-16-family-ovs_datapath
> version: 2.12.0
> license: GPL
> description: Open vSwitch switching datapath
> retpoline: Y
> rhelversion: 7.6
> srcversion: 764C8BD051B3182DE71CF29
> depends:
> nf_conntrack,tunnel6,nf_nat,nf_defrag_ipv6,libcrc32c,nf_nat_ipv6,nf_nat_ipv4
> vermagic: 3.10.0-957.el7.x86_64 SMP mod_unload modversions
> parm: udp_port:Destination UDP port (ushort)
>
> aslo, i uninstall openvswitch-kmod rpm package and use openvswitch.ko in
> linux kernel, like:
> # modinfo openvswitch
> filename:
> /lib/modules/3.10.0-957.el7.x86_64/kernel/net/openvswitch/openvswitch.ko.xz
> alias: net-pf-16-proto-16-family-ovs_packet
> alias: net-pf-16-proto-16-family-ovs_flow
> alias: net-pf-16-proto-16-family-ovs_vport
> alias: net-pf-16-proto-16-family-ovs_datapath
> license: GPL
> description: Open vSwitch switching datapath
> retpoline: Y
> rhelversion: 7.6
> srcversion: 6FE05FC439FA9CE7E264684
> depends:
> nf_conntrack,nf_nat,libcrc32c,nf_nat_ipv6,nf_nat_ipv4,nf_defrag_ipv6
> intree: Y
> vermagic: 3.10.0-957.el7.x86_64 SMP mod_unload modversions
> signer: CentOS Linux kernel signing key
> sig_key: B7:0D:CF:0D:F2:D9:B7:F2:91:59:24:82:49:FD:6F:E8:7B:78:14:27
> sig_hashalgo: sha256
>
> and get the bw is:
> # ip netns exec vm2 iperf3 -c 192.168.100.10 -i 2 -t 10
> Connecting to host 192.168.100.10, port 5201
> [ 4] local 192.168.100.20 port 35270 connected to 192.168.100.10 port 5201
> [ ID] Interval Transfer Bandwidth Retr Cwnd
> [ 4] 0.00-2.00 sec 1.61 GBytes 6.92 Gbits/sec 2793 877 KBytes
> [ 4] 2.00-4.00 sec 1.56 GBytes 6.70 Gbits/sec 7773 907 KBytes
> [ 4] 4.00-6.00 sec 1.78 GBytes 7.62 Gbits/sec 4387 952 KBytes
> [ 4] 6.00-8.00 sec 1.66 GBytes 7.11 Gbits/sec 9365 815 KBytes
> [ 4] 8.00-10.00 sec 1.68 GBytes 7.20 Gbits/sec 2421 554 KBytes
> - - - - - - - - - - - - - - - - - - - - - - - - -
> [ ID] Interval Transfer Bandwidth Retr
> [ 4] 0.00-10.00 sec 8.28 GBytes 7.11 Gbits/sec 26739 sender
> [ 4] 0.00-10.00 sec 8.28 GBytes 7.11 Gbits/sec receiver
>
> iperf Done.
>
> so, the performence also is not normal when use openvswitch.ko installed from
> openvswitch-kmod rpm package.
>
> can you show me your build process for openvswith-*.rpm package or give me
> some link? or your process of install ovs?
> i don't know where is wrong.
The rpm I have try, no reproduces. I think you should check your topo.
> thx.
>
> Tonghao Zhang <[email protected]> 于2019年11月5日周二 下午3:59写道：
>>
>> On Mon, Nov 4, 2019 at 5:14 PM shuangyang qian <[email protected]> wrote:
>> >
>> > Hi:
>> > I make rpm packages for ovs and ovn with this
>> > document:http://docs.openvswitch.org/en/latest/intro/install/fedora/ . For
>> > use the kernel module in ovs tree, i configure with the command:
>> > ./configure --with-linux=/lib/modules/$(uname -r)/build .
>> > Then install the rpm packages.
>> > when i finished, i check the openvswitch.ko is like:
>> > # lsmod | grep openvswitch
>> > openvswitch 291276 0
>> > tunnel6 3115 1 openvswitch
>> > nf_defrag_ipv6 25957 2 nf_conntrack_ipv6,openvswitch
>> > nf_nat_ipv6 6459 2 openvswitch,ip6table_nat
>> > nf_nat_ipv4 6187 2 openvswitch,iptable_nat
>> > nf_nat 18080 5
>> > xt_nat,openvswitch,nf_nat_ipv6,nf_nat_masquerade_ipv4,nf_nat_ipv4
>> > nf_conntrack 102766 10
>> > ip_vs,nf_conntrack_ipv6,openvswitch,nf_conntrack_ipv4,nf_conntrack_netlink,nf_nat_ipv6,nf_nat_masquerade_ipv4,xt_conntrack,nf_nat_ipv4,nf_nat
>> > libcrc32c 1388 3 ip_vs,openvswitch,xfs
>> > ipv6 400397 92
>> > ip_vs,nf_conntrack_ipv6,openvswitch,nf_defrag_ipv6,nf_nat_ipv6,bridge
>> > # modinfo openvswitch
>> > filename:
>> > /lib/modules/4.9.18-19080201/extra/openvswitch/openvswitch.ko
>> > alias: net-pf-16-proto-16-family-ovs_ct_limit
>> > alias: net-pf-16-proto-16-family-ovs_meter
>> > alias: net-pf-16-proto-16-family-ovs_packet
>> > alias: net-pf-16-proto-16-family-ovs_flow
>> > alias: net-pf-16-proto-16-family-ovs_vport
>> > alias: net-pf-16-proto-16-family-ovs_datapath
>> > version: 2.11.2
>> > license: GPL
>> > description: Open vSwitch switching datapath
>> > srcversion: 9DDA327F9DD46B9813628A4
>> > depends:
>> > nf_conntrack,tunnel6,ipv6,nf_nat,nf_defrag_ipv6,libcrc32c,nf_nat_ipv6,nf_nat_ipv4
>> > vermagic: 4.9.18-19080201 SMP mod_unload modversions
>> > parm: udp_port:Destination UDP port (ushort)
>> > # rpm -qf /lib/modules/4.9.18-19080201/extra/openvswitch/openvswitch.ko
>> > openvswitch-kmod-2.11.2-1.el7.x86_64
>> >
>> > Then i start to build my network structure. I have two node，and network
>> > namespace vm1 on node1, network namespace vm2 on node2. vm1's veth pair
>> > veth-vm1 is on node1's br-int. vm2's veth pair veth-vm2 is on node2's
>> > br-int. In logical layer, there is one logical switch test-subnet and two
>> > logical switch port node1 and node2 on it. like this:
>> > # ovn-nbctl show
>> > switch 70585c0e-3cd9-459e-9448-3c13f3c0bfa3 (test-subnet)
>> > port node2
>> > addresses: ["00:00:00:00:00:02 192.168.100.20"]
>> > port node1
>> > addresses: ["00:00:00:00:00:01 192.168.100.10"]
>> > on node1:
>> > # ovs-vsctl show
>> > 5180f74a-1379-49af-b265-4403bd0d82d8
>> > Bridge br-int
>> > fail_mode: secure
>> > Port "ovn-431b9e-0"
>> > Interface "ovn-431b9e-0"
>> > type: geneve
>> > options: {csum="true", key=flow, remote_ip="10.18.124.2"}
>> > Port br-int
>> > Interface br-int
>> > type: internal
>> > Port "veth-vm1"
>> > Interface "veth-vm1"
>> > ovs_version: "2.11.2"
>> > # ip netns exec vm1 ip a
>> > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group
>> > default qlen 1
>> > 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
>> > 14: ovs-gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN
>> > group default qlen 1000
>> > link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
>> > 15: erspan0@NONE: <BROADCAST,MULTICAST> mtu 1450 qdisc noop state DOWN
>> > group default qlen 1000
>> > link/ether 22:02:1b:08:ec:53 brd ff:ff:ff:ff:ff:ff
>> > 16: ovs-ip6gre0@NONE: <NOARP> mtu 1448 qdisc noop state DOWN group default
>> > qlen 1
>> > link/gre6 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00 brd
>> > 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00
>> > 17: ovs-ip6tnl0@NONE: <NOARP> mtu 1452 qdisc noop state DOWN group default
>> > qlen 1
>> > link/tunnel6 :: brd ::
>> > 18: vm1-eth0@if17: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1400 qdisc
>> > noqueue state UP group default qlen 1000
>> > link/ether 00:00:00:00:00:01 brd ff:ff:ff:ff:ff:ff link-netnsid 0
>> > inet 192.168.100.10/24 scope global vm1-eth0
>> > valid_lft forever preferred_lft forever
>> > inet6 fe80::200:ff:fe00:1/64 scope link
>> > valid_lft forever preferred_lft forever
>> >
>> >
>> > on node2:# ovs-vsctl show
>> > 011332d0-78bc-47f7-be3c-fab0beb08e28
>> > Bridge br-int
>> > fail_mode: secure
>> > Port br-int
>> > Interface br-int
>> > type: internal
>> > Port "ovn-c655f8-0"
>> > Interface "ovn-c655f8-0"
>> > type: geneve
>> > options: {csum="true", key=flow, remote_ip="10.18.124.1"}
>> > Port "veth-vm2"
>> > Interface "veth-vm2"
>> > ovs_version: "2.11.2"
>> > #ip netns exec vm2 ip a
>> > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group
>> > default qlen 1
>> > 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
>> > 10: ovs-gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN
>> > group default qlen 1000
>> > link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
>> > 11: erspan0@NONE: <BROADCAST,MULTICAST> mtu 1450 qdisc noop state DOWN
>> > group default qlen 1000
>> > link/ether 4a:1d:ca:65:e3:ca brd ff:ff:ff:ff:ff:ff
>> > 12: ovs-ip6gre0@NONE: <NOARP> mtu 1448 qdisc noop state DOWN group default
>> > qlen 1
>> > link/gre6 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00 brd
>> > 00:00:00:00:00:00:00:00:00:00:00:00:00:00:00:00
>> > 13: ovs-ip6tnl0@NONE: <NOARP> mtu 1452 qdisc noop state DOWN group default
>> > qlen 1
>> > link/tunnel6 :: brd ::
>> > 17: vm2-eth0@if16: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1400 qdisc
>> > noqueue state UP group default qlen 1000
>> > link/ether 00:00:00:00:00:02 brd ff:ff:ff:ff:ff:ff link-netnsid 0
>> > inet 192.168.100.20/24 scope global vm2-eth0
>> > valid_lft forever preferred_lft forever
>> > inet6 fe80::200:ff:fe00:2/64 scope link
>> > valid_lft forever preferred_lft forever
>> >
>> > then i start to use iperf to check the network performence, oh, by the
>> > way, i use geneve protocol between the two nodes, ovn-sbctl show is :
>> > # ovn-sbctl show
>> > Chassis "c655f877-b7ed-4bb5-a047-23521426d541"
>> > hostname: "node1.com"
>> > Encap geneve
>> > ip: "10.18.124.1"
>> > options: {csum="true"}
>> > Port_Binding "node1"
>> > Chassis "431b9efb-b464-42a1-a6dd-7fc6e0176137"
>> > hostname: "node2.com"
>> > Encap geneve
>> > ip: "10.18.124.2"
>> > options: {csum="true"}
>> > Port_Binding "node2"
>> >
>> > on node1, in network namespace vm1 i start the iperf3 as server:
>> > #ip netns exec vm1 iperf3 -s
>> > on node2, in network namespace vm2 i start the iper3 as client:
>> > # ip netns exec vm2 iperf3 -c 192.168.100.10
>> > Connecting to host 192.168.100.10, port 5201
>> > [ 4] local 192.168.100.20 port 40708 connected to 192.168.100.10 port 5201
>> > [ ID] Interval Transfer Bandwidth Retr Cwnd
>> > [ 4] 0.00-1.00 sec 431 MBytes 3.61 Gbits/sec 34 253 KBytes
>> > [ 4] 1.00-2.00 sec 426 MBytes 3.58 Gbits/sec 0 253 KBytes
>> > [ 4] 2.00-3.00 sec 426 MBytes 3.57 Gbits/sec 0 253 KBytes
>> > [ 4] 3.00-4.00 sec 401 MBytes 3.37 Gbits/sec 0 255 KBytes
>> > [ 4] 4.00-5.00 sec 429 MBytes 3.60 Gbits/sec 0 255 KBytes
>> > [ 4] 5.00-6.00 sec 413 MBytes 3.46 Gbits/sec 0 253 KBytes
>> > [ 4] 6.00-7.00 sec 409 MBytes 3.43 Gbits/sec 0 250 KBytes
>> > [ 4] 7.00-8.00 sec 427 MBytes 3.58 Gbits/sec 0 253 KBytes
>> > [ 4] 8.00-9.00 sec 417 MBytes 3.49 Gbits/sec 0 250 KBytes
>> > [ 4] 9.00-10.00 sec 385 MBytes 3.23 Gbits/sec 0 5.27 KBytes
>> > - - - - - - - - - - - - - - - - - - - - - - - - -
>> > [ ID] Interval Transfer Bandwidth Retr
>> > [ 4] 0.00-10.00 sec 4.07 GBytes 3.49 Gbits/sec 34
>> > sender
>> > [ 4] 0.00-10.00 sec 4.07 GBytes 3.49 Gbits/sec
>> > receiver
>> >
>> > as you see, the bw is only 3.xxGbits/sec, but my physics eth1's bw 10000M:
>> Hi, I run the ovs on node1 and node2, using geneve tunnel but I didn't
>> reproduce your issue.
>>
>> create the geneve tunnel:
>> # ovs-vsctl add-br br-int
>> # ovs-vsctl add-port br-int gnv0 -- set Interface gnv0 type=geneve
>> options:csum=true options:key=100 options:remote_ip=1.1.1.200
>>
>> # ovs-vsctl show
>> 9393485c-c64c-490e-884e-418ff5d90251
>> Bridge br-int
>> Port gnv0
>> Interface gnv0
>> type: geneve
>> options: {csum="true", key="100", remote_ip="1.1.1.200"}
>> Port __tap01
>> Interface __tap01
>> Port br-int
>> Interface br-int
>> type: internal
>>
>> # ip netns exec ns100 ifconfig
>> __tap00: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1450
>> inet 2.2.2.100 netmask 255.255.255.0 broadcast 0.0.0.0
>> inet6 fe80::254:ff:fe00:1 prefixlen 64 scopeid 0x20<link>
>> ether 00:54:00:00:00:01 txqueuelen 1000 (Ethernet)
>> RX packets 605000 bytes 39951500 (38.1 MiB)
>> RX errors 0 dropped 0 overruns 0 frame 0
>> TX packets 819864 bytes 31247862764 (29.1 GiB)
>> TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
>>
>> # ip netns exec ns100 iperf -c 2.2.2.200 -i 2 -t 10
>> ------------------------------------------------------------
>> Client connecting to 2.2.2.200, TCP port 5001
>> TCP window size: 482 KByte (default)
>> ------------------------------------------------------------
>> [ 3] local 2.2.2.100 port 41428 connected with 2.2.2.200 port 5001
>> [ ID] Interval Transfer Bandwidth
>> [ 3] 0.0- 2.0 sec 1.85 GBytes 7.93 Gbits/sec
>> [ 3] 2.0- 4.0 sec 1.94 GBytes 8.33 Gbits/sec
>>
>> # modinfo openvswitch
>> filename: /lib/modules/3.10.0-957.1.3.el7.x86_64/extra/openvswitch.ko
>>
>> so, can you use the commands as show above to reproduce your
>> issue.(the kernel version is different.)
>> > # ethtool eth1
>> > Settings for eth1:
>> > Supported ports: [ FIBRE ]
>> > Supported link modes: 10000baseT/Full
>> > Supported pause frame use: Symmetric
>> > Supports auto-negotiation: No
>> > Supported FEC modes: Not reported
>> > Advertised link modes: 10000baseT/Full
>> > Advertised pause frame use: Symmetric
>> > Advertised auto-negotiation: No
>> > Advertised FEC modes: Not reported
>> > Speed: 10000Mb/s
>> > Duplex: Full
>> > Port: Other
>> > PHYAD: 0
>> > Transceiver: external
>> > Auto-negotiation: off
>> > Supports Wake-on: d
>> > Wake-on: d
>> > Current message level: 0x00000007 (7)
>> > drv probe link
>> > Link detected: yes
>> >
>> > when i uninstall the openvswitch-kmod package and use the openvswitch.ko
>> > in the upstream linux kernel, like this:
>> > # lsmod | grep openvswitch
>> > openvswitch 95805 0
>> > nf_defrag_ipv6 25957 2 nf_conntrack_ipv6,openvswitch
>> > nf_nat_ipv6 6459 2 openvswitch,ip6table_nat
>> > nf_nat_ipv4 6187 2 openvswitch,iptable_nat
>> > nf_nat 18080 5
>> > xt_nat,openvswitch,nf_nat_ipv6,nf_nat_masquerade_ipv4,nf_nat_ipv4
>> > nf_conntrack 102766 10
>> > ip_vs,nf_conntrack_ipv6,openvswitch,nf_conntrack_ipv4,nf_conntrack_netlink,nf_nat_ipv6,nf_nat_masquerade_ipv4,xt_conntrack,nf_nat_ipv4,nf_nat
>> > libcrc32c 1388 3 ip_vs,openvswitch,xfs
>> > # modinfo openvswitch
>> > filename:
>> > /lib/modules/4.9.18-19080201/kernel/net/openvswitch/openvswitch.ko
>> > alias: net-pf-16-proto-16-family-ovs_packet
>> > alias: net-pf-16-proto-16-family-ovs_flow
>> > alias: net-pf-16-proto-16-family-ovs_vport
>> > alias: net-pf-16-proto-16-family-ovs_datapath
>> > license: GPL
>> > description: Open vSwitch switching datapath
>> > srcversion: 915B872C96FB1D38D107742
>> > depends:
>> > nf_conntrack,nf_nat,libcrc32c,nf_nat_ipv6,nf_nat_ipv4,nf_defrag_ipv6
>> > intree: Y
>> > vermagic: 4.9.18-19080201 SMP mod_unload modversions
>> >
>> > and do the same test in above, and i get the follow result:
>> > # ip netns exec vm2 iperf3 -c 192.168.100.10
>> > Connecting to host 192.168.100.10, port 5201
>> > [ 4] local 192.168.100.20 port 40652 connected to 192.168.100.10 port 5201
>> > [ ID] Interval Transfer Bandwidth Retr Cwnd
>> > [ 4] 0.00-1.00 sec 1000 MBytes 8.39 Gbits/sec 4 290 KBytes
>> > [ 4] 1.00-2.00 sec 994 MBytes 8.34 Gbits/sec 0 292 KBytes
>> > [ 4] 2.00-3.00 sec 1002 MBytes 8.41 Gbits/sec 0 287 KBytes
>> > 4] 3.00-4.00 sec 994 MBytes 8.34 Gbits/sec 0 292 KBytes
>> > ▽ 4] 4.00-5.00 sec 992 MBytes 8.32 Gbits/sec 0 298 KBytes
>> > [ 4] 5.00-6.00 sec 994 MBytes 8.34 Gbits/sec 0 305 KBytes
>> > [ 4] 6.00-7.00 sec 989 MBytes 8.29 Gbits/sec 0 313 KBytes
>> > [ 4] 7.00-8.00 sec 992 MBytes 8.32 Gbits/sec 0 290 KBytes
>> > [ 4] 8.00-9.00 sec 996 MBytes 8.36 Gbits/sec 0 303 KBytes
>> > [ 4] 9.00-10.00 sec 955 MBytes 8.01 Gbits/sec 0 5.27 KBytes
>> > - - - - - - - - - - - - - - - - - - - - - - - - -
>> > [ ID] Interval Transfer Bandwidth Retr
>> > [ 4] 0.00-10.00 sec 9.67 GBytes 8.31 Gbits/sec 4
>> > sender
>> > [ 4] 0.00-10.00 sec 9.67 GBytes 8.31 Gbits/sec
>> > receiver
>> >
>> > so, i cann't understand why the performence is so poor when i use the
>> > kernel module build from the ovs tree.
>> >
>> > anyone can give me some advice where is wrong?
>> >
>> > thx!
>> >
>> > _______________________________________________
>> > discuss mailing list
>> > [email protected]
>> > https://mail.openvswitch.org/mailman/listinfo/ovs-discuss
_______________________________________________
discuss mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-discuss
