Yang,
I must have missed your email. Sorry about that.

The syntax looks conform to what is specified in the draft to me.

Fabio


From: nvo3 <[email protected]> on behalf of 杨光 <[email protected]>
Date: Sunday, April 12, 2020 at 7:06 PM
To: "[email protected]" <[email protected]>
Subject: [nvo3] Looking help for vxlan-gpe

I hope to get the  some help for the vxlan-gpe protocol.

I am Yang guang from a Cloud computing provider in China,and in charge of the 
load balance service. We plan to use the vxlan-gpe protocol to carry some 
network address information, and cooperate with the NAT function of virtual 
switch,  the real server can see the real client‘s address when the load 
balance using snat+dnat mode.

Here is the header format

 0                   1                   2                   3

 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

|0|0|0|0|I|P|0|0|       Reserved                |   NP = LB 254 |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

|         VXLAN Network Identifier (VNI)        |   Reserved    |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

|   type = v4   |   Length = 3  |   Reserved    |  NP = Eth 03  |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

|                              CIP                              |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

|                              VIP                              |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

|             SPORT             |             DPORT             |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Since vxlan-gpe is still in the draft stage, and there is no sample shim header 
in the draft, I am not sure whether this custom format conforms to the 
regulations and design philosophy. One of my concerns is that the format 
changes in future will be extremely complex if it do not meet the 
specifications.

This mail was sent once before, and the reason to resend is to make sure it was 
delivered successfully, because I received the rejection letter before.

Thank you for your time and consideration
-------
Sincerely Yang guang
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3

Reply via email to