Jesse,
M2/eth0 is physically connected to M3/eth0. Both these machines are running OVS, and have a GRE tunnel configured between them. M2/eth1 is physically connected to M1/eth0. Here M1 is a client machine, whose ping req./resp. are expected to be GRE encapsulated (between M2 & M3). For this purpose, I have configured a flow on M2 - to match all pkts arriving on eth1 port & with M1-eth0-src-mac. to egress via GRE port (gre0). Thanks Rashmi On Tue, Dec 17, 2013 at 10:38 PM, Jesse Gross <je...@nicira.com> wrote: > On Tue, Dec 17, 2013 at 1:00 AM, Rashmi Rashmi <rashmi....@gmail.com> > wrote: > > Hello, > > > > > > > > I am using a GRE tunnel with OVS. Here is the setup detail: > > > > > > > > > > > > +--------------+ eth0 (11.0.0.2) eth1 +----------------+ eth0 > > eth0 +----------------+ > > > > | M1 (Linux) |-------------------------------------| M2 (Linux) > > |-----------------------------------------------| M3 (Linux) | > > > > +--------------+ +----------------+ > p0 > > (11.0.0.1) p0 (11.0.0.3) +----------------+ > > > > > > gre0 <============> gre0 > > > > > > > > I am running OVS (1.9.3) on Machines M2 & M3. > > > > > > > > For GRE tunnel between M2 & M3 – > > > > On pinging 11.0.0.3 from M1, the pkts coming out of M2.eth0 interface are > > GRE encapsulated. But on machine M3, I see that the single ping-request > pkt > > is being looped multiple times. > > What are eth0/1 connected to on M2 and M3? It sounds like unknown > unicast is being flooded over those ports. >
_______________________________________________ discuss mailing list discuss@openvswitch.org http://openvswitch.org/mailman/listinfo/discuss
