Ravi,
You and Jesse don't seem to agree about the MPLS implementation but the QinQ
code doesn't seem to be in conflict. Any chance you would consider spitting the
MPLS and QinQ patches and see if you can get the QinQ one through? I am
already using it live and its working great so would love to
through the switch and observe the load,
hack the kernel path to layer 2 only and replay the same trace again and
compare results.
Cheers,
Chris
On Thu, 29 Mar 2012 15:50:50 Jesse Gross wrote:
> On Wed, Mar 28, 2012 at 4:30 PM, Chris Browning
wrote:
> > First, the kernel module
Hi,
I am a Network Engineer/C Developer for a small Telco in New Zealand. We are
very pro Linux based solutions to problems most telco's would normally throw
lots of cash at big brands for.
The specific problem that started me looking to openvswitch for is this:
We get customers delivered to us
