A basic question about networking

Hello,

I would like to ask a very basic question about general networking.
Please let me know if you think this question should be asked in a
different newsgroup.

I am running VNC server on a linux machine. I have started a VNC
session. Then I connect to this session by running vncviewer from the
same machine. Will all the traffic generated pass on to the network I
am connected to? Or will the TCP/IP stack recognize that the targeted
machine for IP packets is the same as the originating machine and not
send out the packets?

In fact this question can be asked in the context of any general
client/server. Say I run a web server and connect to it from a browser
on the same machine, will the traffic leak onto the network?

Thanks in advance for your responses.

vkj

(E-Mail Removed) wrote:
> Hello,
>
> I would like to ask a very basic question about general networking.
> Please let me know if you think this question should be asked in a
> different newsgroup.
>
> I am running VNC server on a linux machine. I have started a VNC
> session. Then I connect to this session by running vncviewer from the
> same machine. Will all the traffic generated pass on to the network I
> am connected to? Or will the TCP/IP stack recognize that the targeted
> machine for IP packets is the same as the originating machine and not
> send out the packets?
>
> In fact this question can be asked in the context of any general
> client/server. Say I run a web server and connect to it from a browser
> on the same machine, will the traffic leak onto the network?
>
> Thanks in advance for your responses.
>
> vkj


The Linux TCP/IP protocol stack is smart enough to
route packets addressed to any of the local addresses
via the loopback interface, lo.

This means also that you cannot test the Ethernet
interface or NIC driver using self-addressed
packets.

HTH

--

Tauno Voipio
tauno voipio (at) iki fi

Tauno,

Thanks for finding time to reply. However, I have a follow up
question:

> The Linux TCP/IP protocol stack is smart enough to
> route packets addressed to any of the local addresses
> via the loopback interface, lo.

Yes, I remember reading that this is the case with using local
loopback address (e.g. 127.0.0.1). Will the same also apply when
client uses a non-loopback address, say 10.63.3.227 to connect to a
server running on the same machine?

Thanks
vkj

Hello,

(E-Mail Removed) a écrit :
>
> Yes, I remember reading that this is the case with using local
> loopback address (e.g. 127.0.0.1). Will the same also apply when
> client uses a non-loopback address, say 10.63.3.227 to connect to a
> server running on the same machine?

Yes, any local address. For each local address added on an interface a
local route is added in the local routing table. Local routes can be
displayed with the following command :

$ ip route show type local table local

On Oct 30, 2:49 pm, "vee...@gmail.com" <vee...@gmail.com> wrote:

> Yes, I remember reading that this is the case with using local
> loopback address (e.g. 127.0.0.1). Will the same also apply when
> client uses a non-loopback address, say 10.63.3.227 to connect to a
> server running on the same machine?

If it didn't, how would networking work? It's not like there's some
"reflector box" on every network that will send traffic back to you.

DS

David Schwartz <(E-Mail Removed)> wrote:
> On Oct 30, 2:49 pm, "vee...@gmail.com" <vee...@gmail.com> wrote:

> > Yes, I remember reading that this is the case with using local
> > loopback address (e.g. 127.0.0.1). Will the same also apply when
> > client uses a non-loopback address, say 10.63.3.227 to connect to
> > a server running on the same machine?

> If it didn't, how would networking work? It's not like there's some
> "reflector box" on every network that will send traffic back to you.

Although a switch might actually do that if you hand it a frame
destined for your own MAC. And for the non-switched case it would
depend on whether or not the NIC(s) recognized their own MAC(s) on
outbound and mirrored them.

rick jones
--
The glass is neither half-empty nor half-full. The glass has a leak.
The real question is "Can it be patched?"
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...

On Oct 31, 10:43 am, Rick Jones <rick.jon...@hp.com> wrote:

> David Schwartz <dav...@webmaster.com> wrote:

> > If it didn't, how would networking work? It's not like there's some
> > "reflector box" on every network that will send traffic back to you.

> Although a switch might actually do that if you hand it a frame
> destined for your own MAC.

Then the switch is badly broken. Not doing this is required to avoid
infinite loops. Consider two switches each of which convinced that the
other holds a MAC.

DS

>>>>> "veegnu@gmail" == veegnu@gmail com <(E-Mail Removed)> writes:

veegnu@gmail> Yes, I remember reading that this is the case with
veegnu@gmail> using local loopback address (e.g. 127.0.0.1). Will
veegnu@gmail> the same also apply when client uses a non-loopback
veegnu@gmail> address, say 10.63.3.227 to connect to a server
veegnu@gmail> running on the same machine?

Yes. You can see it for yourself. Try to ping 127.0.0.1 and note the
response time. Now, ping 10.63.3.227 and note the response time. Any
difference?

Now, try to ping an another machine attached to the same subnet
(better: attached directly via a cross cable). Do you notice a larger
response time?


--
Lee Sau Dan §õ¦u´° ~{@nJX6X~}

E-mail: (E-Mail Removed)
Home page: http://www.informatik.uni-freiburg.de/~danlee