[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openvswitch-discuss
Subject:    [ovs-discuss] [dpdk-ovs]VXLAN encapsulation exceeds the MTU of dpdk port .
From:       gowrishankar.m () linux ! vnet ! ibm ! com (gowrishankar)
Date:       2015-06-30 15:13:40
Message-ID: 5592AF54.10803 () linux ! vnet ! ibm ! com
[Download RAW message or body]

On Thursday 25 June 2015 01:31 PM, Dongjun wrote:
> As following topology, two VMs communicate via VXLAN tunnel.
> TCP pkts may be droped for exceeding the MTU of host DPDK port in br2.
> Now I can change the VMs' MTU from default to smaller to accommodate 
> the traffics, it works well.
>
> The way I have not find to change mtu of DPDK phy port, is there a 
> blueprint for supporting this by ovs, like "ovs-appctl 
> netdev-dpdk/set-xxx" or another way?
>
>         Host1                  |                     Host2
>                     (l-ip)-   L3    -(r-ip)
> -----   -----        -----     |     -----       -----  ----- -----
> |VM1| --|br1|Vxlan --|br2|DPDK---DPDK|br2|--Vxlan|br2|--|br1|--|VM2|
> -----   -----        -----     |     -----       -----  ----- -----

There are two br2 bridge on Host2. It is intentional ?

Btw, I am trying on same page but could not simplify the vm-vxlan-dpdk 
working within "single" bridge case,
may be that I am yet to find right flow rules to handle vhost-user, 
vxlan and dpdk ports.
I hope you are using patch port between two bridges. Would you really 
need it ?


Regards,
Gowri Shankar
> _______________________________________________
> discuss mailing list
> discuss at openvswitch.org
> http://openvswitch.org/mailman/listinfo/discuss



[prev in list] [next in list] [prev in thread] [next in thread]