On Tue, Nov 06, 2018 at 02:09:23PM -0500, Siva Teja ARETI wrote: > Answers in line. > > Siva Teja. > > On Tue, Nov 6, 2018 at 1:56 PM Flavio Leitner <f...@sysclose.org> wrote: > > > On Tue, Nov 06, 2018 at 11:51:49AM -0500, Siva Teja ARETI wrote: > > > Hi Greg, > > > > > > Thanks for looking into this. > > > > > > I have two VMs in my setup each with two interfaces. Trying to setup the > > > VXLAN tunnels across these interfaces which are in different subnets. A > > > docker container is attached to ovs bridge using ovs-docker utility on > > each > > > VM and doing a ping from one container to another. > > > > Do you see any interesting related messages in 'dmesg' output or in > > ovs-vswitchd.log? > > > > I could not find any interesting messages in dmesg or in ovs-vswitchd.log > output. > > > > If I recall correctly, the "ip l" should show the vxlan dev named > > vxlan_sys_<port> > > > > Yes. I can see the dev on both of my VMs > > [root@vm1 ~]# ifconfig vxlan_sys_4789 > vxlan_sys_4789: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 65000 > inet6 fe80::2a:28ff:fed2:d4f6 prefixlen 64 scopeid 0x20<link> > ether 02:2a:28:d2:d4:f6 txqueuelen 1000 (Ethernet) > RX packets 0 bytes 0 (0.0 B) > RX errors 0 dropped 0 overruns 0 frame 0 > TX packets 48 bytes 1680 (1.6 KiB) > TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
Do you see TX increasing as you execute the test? or in ovs-ofctl dump-ports <bridge> ? Thanks, fbl > > > > > fbl > > > > > > > > *VM1 details:* > > > > > > [root@vm1 ~]# ip a > > > ....... > > > 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > > state > > > UP qlen 1000 > > > link/ether 52:54:00:b8:05:be brd ff:ff:ff:ff:ff:ff > > > inet 30.30.0.59/24 brd 30.30.0.255 scope global dynamic eth1 > > > valid_lft 3002sec preferred_lft 3002sec > > > inet6 fe80::5054:ff:feb8:5be/64 scope link > > > valid_lft forever preferred_lft forever > > > 4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > > state > > > UP qlen 1000 > > > link/ether 52:54:00:f0:64:37 brd ff:ff:ff:ff:ff:ff > > > inet 20.20.0.183/24 brd 20.20.0.255 scope global dynamic eth2 > > > valid_lft 3248sec preferred_lft 3248sec > > > inet6 fe80::5054:ff:fef0:6437/64 scope link > > > valid_lft forever preferred_lft forever > > > ....... > > > [root@vm1 ~]# ovs-vsctl show > > > ff70c814-d1b0-4018-aee8-8b635187afee > > > Bridge "testbr0" > > > Port "gre0" > > > Interface "gre0" > > > type: gre > > > options: {local_ip="20.20.0.183", > > remote_ip="30.30.0.193"} > > > Port "testbr0" > > > Interface "testbr0" > > > type: internal > > > Port "2cfb62a9b0f04_l" > > > Interface "2cfb62a9b0f04_l" > > > ovs_version: "2.9.2" > > > [root@vm1 ~]# ip rule list > > > 0: from all lookup local > > > 32765: from 20.20.0.183 lookup siva > > > 32766: from all lookup main > > > 32767: from all lookup default > > > [root@vm1 ~]# ip route show table siva > > > default dev eth2 scope link src 20.20.0.183 > > > [root@vm1 ~]# ######################### A docker container is attached > > to > > > ovs bridge using ovs-docker utility > > > [root@vm1 ~]# docker ps > > > CONTAINER ID IMAGE COMMAND CREATED > > > STATUS PORTS NAMES > > > be4ab434db99 busybox "sh" 5 days ago > > > Up 5 days admiring_euclid > > > [root@vm1 ~]# nsenter -n -t `docker inspect be4 > > --format={{.State.Pid}}` -- > > > ip a > > > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 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 > > > 2: gre0@NONE: <NOARP> mtu 1476 qdisc noop state DOWN qlen 1 > > > link/gre 0.0.0.0 brd 0.0.0.0 > > > 3: gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN > > qlen > > > 1000 > > > link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff > > > 9: eth0@if10: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue > > > state UP qlen 1000 > > > link/ether 22:98:41:0f:e8:50 brd ff:ff:ff:ff:ff:ff link-netnsid 0 > > > inet 70.70.0.10/24 scope global eth0 > > > valid_lft forever preferred_lft forever > > > inet6 fe80::2098:41ff:fe0f:e850/64 scope link > > > valid_lft forever preferred_lft forever > > > > > > > > > *VM2 details:* > > > > > > [root@vm2 ~]# ip a > > > ........ > > > 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > > state > > > UP qlen 1000 > > > link/ether 52:54:00:79:ef:92 brd ff:ff:ff:ff:ff:ff > > > inet 30.30.0.193/24 brd 30.30.0.255 scope global dynamic eth1 > > > valid_lft 2406sec preferred_lft 2406sec > > > inet6 fe80::5054:ff:fe79:ef92/64 scope link > > > valid_lft forever preferred_lft forever > > > 4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > > state > > > UP qlen 1000 > > > link/ether 52:54:00:05:93:7c brd ff:ff:ff:ff:ff:ff > > > inet 20.20.0.64/24 brd 20.20.0.255 scope global dynamic eth2 > > > valid_lft 2775sec preferred_lft 2775sec > > > inet6 fe80::5054:ff:fe05:937c/64 scope link > > > valid_lft forever preferred_lft forever > > > ....... > > > [root@vm2 ~]# ovs-vsctl show > > > b85514db-3f29-4f7a-9001-37d70adfca34 > > > Bridge "testbr0" > > > Port "gre0" > > > Interface "gre0" > > > type: gre > > > options: {local_ip="30.30.0.193", > > remote_ip="20.20.0.183"} > > > Port "a0769422cfc04_l" > > > Interface "a0769422cfc04_l" > > > Port "testbr0" > > > Interface "testbr0" > > > type: internal > > > ovs_version: "2.9.2" > > > [root@vm2 ~]# ip rule list > > > 0: from all lookup local > > > 32766: from all lookup main > > > 32767: from all lookup default > > > [root@vm2 ~]# ######################### A docker container is attached > > to > > > ovs bridge using ovs-docker utility > > > [root@vm2 ~]# docker ps > > > CONTAINER ID IMAGE COMMAND CREATED > > > STATUS PORTS NAMES > > > 86214f0d99e8 busybox:latest "sh" 5 days ago > > > Up 5 days peaceful_snyder > > > [root@vm2 ~]# nsenter -n -t `docker inspect 862 > > --format={{.State.Pid}}` -- > > > ip a > > > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 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 > > > 2: gre0@NONE: <NOARP> mtu 1476 qdisc noop state DOWN qlen 1 > > > link/gre 0.0.0.0 brd 0.0.0.0 > > > 3: gretap0@NONE: <BROADCAST,MULTICAST> mtu 1462 qdisc noop state DOWN > > qlen > > > 1000 > > > link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff > > > 9: eth0@if10: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue > > > state UP qlen 1000 > > > link/ether ae:ac:14:7a:40:5f brd ff:ff:ff:ff:ff:ff link-netnsid 0 > > > inet 70.70.0.20/24 scope global eth0 > > > valid_lft forever preferred_lft forever > > > inet6 fe80::acac:14ff:fe7a:405f/64 scope link > > > valid_lft forever preferred_lft forever > > > > > > With this configuration, if I do a ping from docker container on VM1 to > > > docker container on VM2 it works. > > > > > > [root@vm1 ~]# nsenter -n -t `docker inspect be4 > > --format={{.State.Pid}}` -- > > > ping 70.70.0.20 > > > PING 70.70.0.20 (70.70.0.20) 56(84) bytes of data. > > > 64 bytes from 70.70.0.20: icmp_seq=1 ttl=64 time=0.831 ms > > > 64 bytes from 70.70.0.20: icmp_seq=2 ttl=64 time=0.933 ms > > > 64 bytes from 70.70.0.20: icmp_seq=3 ttl=64 time=0.564 ms > > > ^C > > > --- 70.70.0.20 ping statistics --- > > > 3 packets transmitted, 3 received, 0% packet loss, time 2001ms > > > rtt min/avg/max/mdev = 0.564/0.776/0.933/0.155 ms > > > > > > And the traffic is as expected on VM2. > > > > > > [root@vm2 ~]# tcpdump -n -i any host 20.20.0.183 > > > > > > > > > tcpdump: verbose output suppressed, use -v or -vv for full protocol > > decode > > > listening on any, link-type LINUX_SLL (Linux cooked), capture size 262144 > > > bytes > > > 16:37:32.262553 IP 20.20.0.183 > 30.30.0.193: GREv0, length 102: IP > > > 70.70.0.10 > 70.70.0.20: ICMP echo request, id 28158, seq 1, length 64 > > > 16:37:32.262835 IP 30.30.0.193 > 20.20.0.183: GREv0, length 102: IP > > > 70.70.0.20 > 70.70.0.10: ICMP echo reply, id 28158, seq 1, length 64 > > > 16:37:33.263211 IP 20.20.0.183 > 30.30.0.193: GREv0, length 102: IP > > > 70.70.0.10 > 70.70.0.20: ICMP echo request, id 28158, seq 2, length 64 > > > 16:37:33.263374 IP 30.30.0.193 > 20.20.0.183: GREv0, length 102: IP > > > 70.70.0.20 > 70.70.0.10: ICMP echo reply, id 28158, seq 2, length 64 > > > 16:37:34.264159 IP 20.20.0.183 > 30.30.0.193: GREv0, length 102: IP > > > 70.70.0.10 > 70.70.0.20: ICMP echo request, id 28158, seq 3, length 64 > > > 16:37:34.264252 IP 30.30.0.193 > 20.20.0.183: GREv0, length 102: IP > > > 70.70.0.20 > 70.70.0.10: ICMP echo reply, id 28158, seq 3, length 64 > > > > > > But when I change the tunnel type to vxlan, ping fails. > > > > > > [root@vm1 ~]# ovs-vsctl del-port testbr0 gre0 > > > [root@vm1 ~]# ovs-vsctl add-port testbr0 vxlan0 -- set interface vxlan0 > > > type=vxlan options:local_ip=20.20.0.183 options:remote_ip=30.30.0.193 > > > options:dst_port=4789 > > > [root@vm1 ~]# ovs-vsctl show > > > ff70c814-d1b0-4018-aee8-8b635187afee > > > Bridge "testbr0" > > > Port "testbr0" > > > Interface "testbr0" > > > type: internal > > > Port "vxlan0" > > > Interface "vxlan0" > > > type: vxlan > > > options: {dst_port="4789", local_ip="20.20.0.183", > > > remote_ip="30.30.0.193"} > > > Port "2cfb62a9b0f04_l" > > > Interface "2cfb62a9b0f04_l" > > > ovs_version: "2.9.2" > > > > > > [root@vm2 ~]# ovs-vsctl del-port testbr0 gre0 > > > [root@vm2 ~]# ovs-vsctl add-port testbr0 vxlan0 -- set interface vxlan0 > > > type=vxlan options:local_ip=30.30.0.193 options:remote_ip=20.20.0.183 > > > options:dst_port=4789 > > > [root@vm2 ~]# ovs-vsctl show > > > b85514db-3f29-4f7a-9001-37d70adfca34 > > > Bridge "testbr0" > > > Port "a0769422cfc04_l" > > > Interface "a0769422cfc04_l" > > > Port "vxlan0" > > > Interface "vxlan0" > > > type: vxlan > > > options: {dst_port="4789", local_ip="30.30.0.193", > > > remote_ip="20.20.0.183"} > > > Port "testbr0" > > > Interface "testbr0" > > > type: internal > > > ovs_version: "2.9.2" > > > > > > Ping fails with this setup > > > > > > [root@vm1 ~]# nsenter -n -t `docker inspect be4 > > --format={{.State.Pid}}` -- > > > ping 70.70.0.20 > > > PING 70.70.0.20 (70.70.0.20) 56(84) bytes of data. > > > ^C > > > --- 70.70.0.20 ping statistics --- > > > 6 packets transmitted, 0 received, 100% packet loss, time 4999ms > > > > > > Expected traffic is not seen on VM2 > > > > > > [root@vm2 ~]# tcpdump -n -i any host 20.20.0.183 > > > > > > > > > tcpdump: verbose output suppressed, use -v or -vv for full protocol > > decode > > > listening on any, link-type LINUX_SLL (Linux cooked), capture size 262144 > > > bytes > > > ^C > > > 0 packets captured > > > 0 packets received by filter > > > 0 packets dropped by kernel > > > > > > Kindly let me know if you need more information. > > > > > > Siva Teja. > > > > > > On Tue, Nov 6, 2018 at 10:49 AM Gregory Rose <gvrose8...@gmail.com> > > wrote: > > > > > > > > > > > On 11/5/2018 6:10 PM, Siva Teja ARETI wrote: > > > > > > > > Hi, > > > > > > > > I am trying to use local_ip option for a VXLAN tunnel using ovs but it > > > > does not seem to work. The same works when I use GRE tunnel. I also > > found a > > > > previous discussion from another user who tried the exact same > > approach. > > > > Here is the link to the discussion > > > > > > > > * > > https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg03643.html > > > > < > > https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg03643.html>* > > > > > > > > I am unable to find any working resolution at the end of this > > discussion. > > > > Could you please help? > > > > > > > > > > > > I looked into that but was never able to set up a configuration like > > the > > > > one in that discussion and could > > > > not repro the bug. > > > > > > > > Please provide some details on your usage, configuration and steps to > > > > repro and I can look into it. > > > > > > > > Thanks, > > > > > > > > - Greg > > > > > > > > > > > > I am using ovs 2.9.2 > > > > > > > > [root@localhost ~]# ovs-vsctl --version > > > > ovs-vsctl (Open vSwitch) 2.9.2 > > > > DB Schema 7.15.1 > > > > > > > > Thanks, > > > > Siva Teja. > > > > > > > > > > > > _______________________________________________ > > > > discuss mailing listdiscuss@openvswitch.orghttps:// > > mail.openvswitch.org/mailman/listinfo/ovs-discuss > > > > > > > > > > > > > > > > > _______________________________________________ > > > discuss mailing list > > > disc...@openvswitch.org > > > https://mail.openvswitch.org/mailman/listinfo/ovs-discuss > > > > > > -- > > Flavio > > > > -- Flavio _______________________________________________ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss