3 Replies Latest reply on Aug 3, 2012 7:01 PM by Patrick_Kutch

    VM-to-VM communication Failed even using VFs from the same PF

    Weirong

      Hi,

       

        I'm following the Figure 16 of the Whitepaper: Configuring QoS Features with Intel* Flexible Port Partitioning

      [root@syst-intel-linux-01 ~]# ip link show

      13: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000

          link/ether 00:1b:21:a2:73:8c brd ff:ff:ff:ff:ff:ff

          vf 0 MAC 0a:a2:f3:e4:f6:49

          vf 1 MAC 62:21:3d:b6:46:63

       

      I assigned eth1.vf0 to VM1 and eth1.vf1 to VM2.

      In VM1, I have:

      eth0 Link encap:Ethernet  HWaddr 0A:A2:F3:E4:F6:49
      inet addr:192.1.1.1  Bcast:192.1.1.255  Mask:255.255.255.0

      In VM2, I have:

      eth0 Link encap:Ethernet  HWaddr 62:21:3D:B6:46:63
      inet addr:192.1.1.2  Bcast:192.1.1.255  Mask:255.255.255.0

      In Dom0,

      [root@syst-intel-linux-01 ~]# ip link set eth1 vf 0 vlan 1234

      [root@syst-intel-linux-01 ~]# ip link set eth1 vf 1 vlan 1234

       

      I tried pinging from 192.1.1.1 in VM1 to 192.1.1.2 in VM2. But NO success.

      "tcpdump" on eth0@VM1 showed:

      ...

      13:21:31.596049 ARP, Request who-has 192.1.1.2 tell 192.1.1.1, length 28

      ...

      "tcpdump" on eth0@VM2 showed nothing was received.

       

      Fedora14 was used for both Dom0 and VMs. The Hypervisor was KVM.

       

      Any help will be appreciated. Thanks.