TCP & Existing connections

[karthikbalaguru]

Hi,
Why TCP does not support the functionality of taking
over of the Existing connections by the other interface ?

Thx in advans,
Karthik Balaguru

[Rick Jones]

In comp.protocols.tcp-ip karthikbalaguru <karthikb...@gmail.com> wrote:
> Why TCP does not support the functionality of taking
> over of the Existing connections by the other interface ?

It isn't entirely clear what you are asking. That said:

A TCP connection is uniquely identified by the four-tuple of
local/remote IP and local/remote port number. If the four-tuple is
different, it is a different TCP connection.

Dealing with things like interface failure is handled at layer2 and
layer3 in the 7 (or 9 if you include financial and political) layer
model. When handled at layer two, you will see terms like LACP, bond,
trunk, aggregate and team in your web searches. When handled at layer
three (eg IP) it is called "routing." In the case of IP, there is
also the more fundamental question of whether an IP address belongs to
the system or to the interface - you will see terms like "weak
end-system" and "strong end-system" model when you do web searches.

So, TCP does not support dealing with link/interface failure because
it does not need to - those things are dealt with elsewhere. For TCP
to deal with it would require a TCP connection to be named via
something other than the four-tuple of local/remote IP, local/remote
port number, complicating it unnecessarily, when IP and the data-link
can deal with that to the benefit of more than just TCP. (eg UDP,
etc...)

rick jones
--
The computing industry isn't as much a game of "Follow The Leader" as
it is one of "Ring Around the Rosy" or perhaps "Duck Duck Goose."
- Rick Jones
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...

[David Schwartz]

On Nov 9, 11:32 am, karthikbalaguru <karthikbalagur...@gmail.com>
wrote:

> Why TCP does not support the functionality of taking
> over of the Existing connections by the other interface ?

I'm making some guesses about what you mean, but assuming you mean
what I think you mean, ...

Why doesn't TCP provide a graphical interface and steer your
lawnmower? That's just not its job. If you want/need that capability,
you are welcome to implement it on top of TCP.

DS

[Burkhard Ott]

On Mon, 09 Nov 2009 12:45:06 -0800, David Schwartz wrote:

> On Nov 9, 11:32 am, karthikbalaguru <karthikbalagur...@gmail.com> wrote:
>
>> Why TCP does not support the functionality of taking over of the
>> Existing connections by the other interface ?

> That's just not its job. If you want/need that capability, you are

> welcome to implement it on top of TCP.
>
> DS

or below (CARP, vrrp etc.)

cheers

[Maxwell Lol]

karthikbalaguru <karthikb...@gmail.com> writes:

> Hi,
> Why TCP does not support the functionality of taking
> over of the Existing connections by the other interface ?

Read Rick's answer. But here's something to consider
Would you like someone else to steal your connection without your permission?

[Rick Jones]

In comp.protocols.tcp-ip Rick Jones <rick....@hp.com> wrote:
> ... is handled at layer2 and layer3 in the 7 (or 9 if you include
> financial and political) layer model...

having mentioned the 9 layer model I should probably provide a pointer
to an instance of it:

https://www.isc.org/files/9layer.thumb.png

rick jones
--
denial, anger, bargaining, depression, acceptance, rebirth...
where do you want to be today?

[EJP]

karthikbalaguru wrote:
>
> Why TCP does not support the functionality of taking
> over of the Existing connections by the other interface ?

Should it? Why? Have you thought through the security implications of
that? Do you have a specific proposal in mind that will preserve
security? Do you also have a realistic use-case? or is this just an
academic question?