The term “complexity” is not applicable to this discussion. Even if there were
a priori knowledge that the encapsulated packet is an IPvX packet, the first
nibble should be checked to confirm that it actually encodes the value 6/4
for sanity checking purposes. Complexity would be if there were
On Thu, May 17, 2018 at 11:12 PM, 徐小虎(义先) wrote:
> Tom,
>
> It seems that your argument is based on an assumption that GUE is the only
> UDP tunnel mechanism implemented on the chip. What about if other UDP tunnel
> mechanisms (e.g., VXLAN, GENEVE and MPLS-in-UDP) are
Tom,
It seems that your argument is based on an assumption that GUE is the only UDP
tunnel mechanism implemented on the chip. What about if other UDP tunnel
mechanisms (e.g., VXLAN, GENEVE and MPLS-in-UDP) are implemented as well?
Best regards,Xiaohu
On Thu, May 17, 2018 at 6:41 PM, 徐小虎(义先) wrote:
> It's different. In GUE variant 1, the tunnel decapsulation device needs to
> read the UDP destination port to determine that the UDP payload is a GUE and
> then to read the first two nibble of the GUE payload to
It's different. In GUE variant 1, the tunnel decapsulation device needs to read
the UDP destination port to determine that the UDP payload is a GUE and then to
read the first two nibble of the GUE payload to determine whether the UDP
payload is actually an IP payload. It introduces an
>Because it isn’t different. Again, see GUE variant 1.
Correct. There is no reason to progress another draft that does the
same thing as GUE variant 1.
Fred
From: Int-area [mailto:int-area-boun...@ietf.org] On Behalf Of Joe Touch
Sent: Thursday, May 17, 2018 7:55 AM
To: sarik...@ieee.org
Cc:
You can also carry IP over UDP by using MPLS over IP, and including an explicit
null as the only label in the label stack.
A node that did not want to know about MPLS could treat the label as opaque
data.
- Stewart
___
Int-area mailing list
On Thu, May 17, 2018 at 7:18 AM, Behcet Sarikaya wrote:
>
>
> On Wed, May 16, 2018 at 10:22 PM, 徐小虎(义先)
> wrote:
>>
>> It doesn't matter whether or not it's already there. IMHO, given the
>> popularity of different overlay technologies such as
Because it isn’t different. Again, see GUE variant 1.
> On May 17, 2018, at 7:18 AM, Behcet Sarikaya wrote:
>
>
>
>> On Wed, May 16, 2018 at 10:22 PM, 徐小虎(义先) wrote:
>> It doesn't matter whether or not it's already there. IMHO, given the
On Wed, May 16, 2018 at 10:22 PM, 徐小虎(义先)
wrote:
> It doesn't matter whether or not it's already there. IMHO, given the
> popularity of different overlay technologies such as VXLAN and MPLS-in-UDP
> in practice, GUE initially and mainly targeted as a DC overlay
What overhead?
Look at variant 1.
IMO, it is not useful to take on a new WG item that directly competes with an
existing one that is a superset already.
Joe
> On May 16, 2018, at 8:22 PM, 徐小虎(义先) wrote:
>
> It doesn't matter whether or not it's already there.
Agree with Joe and Tom.
From: Int-area [mailto:int-area-boun...@ietf.org] On Behalf Of Joe Touch
Sent: Tuesday, May 15, 2018 11:47 PM
To: "徐小虎(义先)"
Cc: Internet Area ; intarea-chairs
; draft-xu-intarea-ip-in-udp
It’s not complex. It’s already there. So there continues to be no reason to
waste either a port number or further time discussing this draft.
Joe
> On May 15, 2018, at 9:01 PM, 徐小虎(义先) wrote:
>
> IMHO,there seems no need to introduce such complexity into GUE just
13 matches
Mail list logo