I didn’t try this, so I haven’t any details, will opnfv sfc have such use case?
The below command can check if the packet is dropped or not sudo ovs-appctl dpif/dump-flows br-int From: sfc-dev-boun...@lists.opendaylight.org [mailto:sfc-dev-boun...@lists.opendaylight.org] On Behalf Of Brady Allen Johnson Sent: Thursday, March 2, 2017 11:48 PM To: opnfv-tech-discuss@lists.opnfv.org; srikanth.ling...@nxp.com; sfc-...@lists.opendaylight.org Subject: Re: [sfc-dev] [SFC] ODL SFC with multiple compute nodes Srikanth, We experienced a similar problem lately in OPNFV SFC when using multiple compute hosts. The situation was that the packets returning from an SF entered the SFF on the XVLAN-GPE port, and then the packets need to go to an SFF on a different compute host, and they do so on a normal VXLAN tunnel, which is obviously on a different port than the VXLAN-GPE port. The problem is that when the packets "transition" from the VXLAN-GPE tunnel to the VXLAN tunnel, the UDP port number in the packet was not correctly updated, and it was still 6633 from VXLAN-GPE and got dropped by the VXLAN port. Can you do a tcpdump on the first compute host to see if you are experiencing the same problem? I believe this was fixed, but I cant remember where the solution is. Hopefully Yi Yang can provide details about this. Regards, Brady -----Original Message----- From: Brady Allen Johnson <brady.allen.john...@ericsson.com<mailto:brady%20allen%20johnson%20%3cbrady.allen.john...@ericsson.com%3e>> To: opnfv-tech-discuss@lists.opnfv.org<mailto:opnfv-tech-discuss@lists.opnfv.org> <opnfv-tech-discuss@lists.opnfv.org<mailto:%22opnfv-tech-disc...@lists.opnfv.org%22%20%3copnfv-tech-disc...@lists.opnfv.org%3e>>, srikanth.ling...@nxp.com<mailto:srikanth.ling...@nxp.com> <srikanth.ling...@nxp.com<mailto:%22srikanth.ling...@nxp.com%22%20%3csrikanth.ling...@nxp.com%3e>>, sfc-...@lists.opendaylight.org<mailto:sfc-...@lists.opendaylight.org> <sfc-...@lists.opendaylight.org<mailto:%22sfc-...@lists.opendaylight.org%22%20%3csfc-...@lists.opendaylight.org%3e>> Subject: [sfc-dev] [SFC] ODL SFC with multiple compute nodes Date: Thu, 2 Mar 2017 13:42:16 +0000 Since this is more of an issue with OPNFV SFC than ODL SFC, I will add that mail list. Regards, Brady -----Original Message----- From: Srikanth Lingala <srikanth.ling...@nxp.com<mailto:srikanth%20lingala%20%3csrikanth.ling...@nxp.com%3e>> To: sfc-...@lists.opendaylight.org<mailto:sfc-...@lists.opendaylight.org> <sfc-...@lists.opendaylight.org<mailto:%22sfc-...@lists.opendaylight.org%22%20%3csfc-...@lists.opendaylight.org%3e>> Subject: [sfc-dev] ODL SFC with multiple compute nodes Date: Thu, 2 Mar 2017 08:41:27 +0000 PS: Sorry, If I am resending the mail to the sfc-dev mailing list. As I can’t find the mail in the mailing list, I am resending it. Hi, I am able to run SFC usecase successfully with single Compute Node. Now, I am trying SFC usecase across multiple computes with Opendaylight, Openstack and OVS with NSH patches by Yi Yang. For that, I deployed a Openstack setup with OPNFV Colorado 3.0. My setup includes: · One Openstack Controller with ODL controller and tacker · Two Compute Nodes For validating SFC usecase, I am using tacker. Following are the sequence of commands to configure SFC attributes, in my usecase: neutron net-create net_mgmt --provider:network_type=vxlan --provider:segmentation_id 1005 neutron subnet-create net_mgmt 123.123.123.0/24 tacker vnfd-create --vnfd-file /root/sfc-random/test-vnfd-1.yaml tacker vnfd-create --vnfd-file /root/sfc-random/test-vnfd-2.yaml tacker vnf-create --name testVNF1 --vnfd-name test-vnfd-1 (SFC VM1 on Compute 1) tacker vnf-create --name testVNF2 --vnfd-name test-vnfd-2 (SFC VM2 on Compute 2) nova boot --flavor custom --image SFC-VM --nic net-id=$(neutron net-list | awk '/ net_mgmt / {print $2}') --availability-zone compute1 http_server (http_server VM on Compute 2) nova boot --flavor custom --image SFC-VM --nic net-id=$(neutron net-list | awk '/ net_mgmt / {print $2}') --availability-zone compute2 http_client (http_client VM on Compute 1) tacker sfc-create --name mychain --chain testVNF1,testVNF2 tacker sfc-classifier-create --name myclass --chain mychain --match source_port=2000,dest_port=80,protocol=6 All the above commands are successful. SFC flows are added in the respective compute nodes. Flows on Compute Node1: http://pastebin.com/R1ZkvpHt Flows on Compute Node2: http://pastebin.com/WcCLeQTH Here comes the issue…. I am able to see packets on SFC Flows in Compute Node1. And also, I am able to see VxLAN packets (vxlan_tool.py) on SFC VM launched on Compute Node 1. But, I am not able to see any packets on SFC flows in Compute Node 2. Obviously, not getting VxLAN packets on SFC VM in Compute Node2. I think packets are not getting from ‘vxgpe’ port in Compute Node1 to Compute Node 2. When I execute the command ovs-appctl ofproto/trace on br-int of Compute Node1, I got the following trace information: Here, vxgpe port number is 1. So, I gave flow trace for in_port as ‘1’. Compute_Node1#> ovs-appctl ofproto/trace br-int nsi=254,nsp=35,in_port=1 Bridge: br-int Flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x0000 Rule: table=0 cookie=0x14 priority=250,nsp=35 OpenFlow actions=goto_table:152 Resubmitted flow: in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x0000 Resubmitted regs: reg0=0x0 reg1=0x0 reg2=0x0 reg3=0x0 reg4=0x0 reg5=0x0 reg6=0x0 reg7=0x0 reg8=0x0 reg9=0x0 reg10=0x0 reg11=0x0 reg12=0x0 reg13=0x0 reg14=0x0 reg15=0x0 Resubmitted odp: drop Resubmitted megaflow: recirc_id=0,nsi=254,nsp=35,in_port=1,dl_type=0x0000 Rule: table=152 cookie=0x14 priority=550,nsi=254,nsp=35 OpenFlow actions=load:0xc0a8001a->NXM_NX_TUN_IPV4_DST[],goto_table:158 Resubmitted flow: tun_src=0.0.0.0,tun_dst=192.168.0.26,tun_ipv6_src=::,tun_ipv6_dst=::,tun_gbp_id=0,tun_gbp_flags=0,tun_gpe_np=0,tun_gpe_flags=0,tun_tos=0,tun_ttl=0,tun_flags=0,in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x0000 Resubmitted regs: reg0=0x0 reg1=0x0 reg2=0x0 reg3=0x0 reg4=0x0 reg5=0x0 reg6=0x0 reg7=0x0 reg8=0x0 reg9=0x0 reg10=0x0 reg11=0x0 reg12=0x0 reg13=0x0 reg14=0x0 reg15=0x0 Resubmitted odp: drop Resubmitted megaflow: recirc_id=0,nsi=254,nsp=35,tun_dst=0.0.0.0,in_port=1,dl_type=0x0000 Rule: table=158 cookie=0xba5eba1100000101 priority=655,nsi=254,nsp=35,in_port=1 OpenFlow actions=move:NXM_NX_NSH_MDTYPE[]->NXM_NX_NSH_MDTYPE[],move:NXM_NX_NSH_NP[]->NXM_NX_NSH_NP[],move:NXM_NX_NSH_C1[]->NXM_NX_NSH_C1[],move:NXM_NX_NSH_C2[]->NXM_NX_NSH_C2[],move:NXM_NX_TUN_ID[0..31]->NXM_NX_TUN_ID[0..31],load:0x4->NXM_NX_TUN_GPE_NP[],IN_PORT output to native tunnel native tunnel routing failed Final flow: tun_src=0.0.0.0,tun_dst=192.168.0.26,tun_ipv6_src=::,tun_ipv6_dst=::,tun_gbp_id=0,tun_gbp_flags=0,tun_gpe_np=0x4,tun_gpe_flags=0,tun_tos=0,tun_ttl=0,tun_flags=0,in_port=1,vlan_tci=0x0000,dl_src=00:00:00:00:00:00,dl_dst=00:00:00:00:00:00,dl_type=0x0000 Megaflow: recirc_id=0,nsi=254,nsh_mdtype=0,nsh_np=0,nsp=35,nshc1=0,nshc2=0,tun_id=0/0xffffffff,tun_dst=0.0.0.0,tun_gpe_np=0,in_port=1,dl_type=0x0000 Datapath actions: drop Please help me to resolve the issue. Regards, Srikanth. _______________________________________________ sfc-dev mailing list sfc-...@lists.opendaylight.org<mailto:sfc-...@lists.opendaylight.org> https://lists.opendaylight.org/mailman/listinfo/sfc-dev _______________________________________________ sfc-dev mailing list sfc-...@lists.opendaylight.org<mailto:sfc-...@lists.opendaylight.org> https://lists.opendaylight.org/mailman/listinfo/sfc-dev
_______________________________________________ opnfv-tech-discuss mailing list opnfv-tech-discuss@lists.opnfv.org https://lists.opnfv.org/mailman/listinfo/opnfv-tech-discuss
