Re: network delay loop ?

--===============3595397165983652676==
Content-Type: multipart/alternative; boundary="000000000000b449c2057ebee4c9"

--000000000000b449c2057ebee4c9
Content-Type: text/plain; charset="UTF-8"

Heya Sam, I still have that Network Nightmare box, it is sitting on a shelf
in my basement.  The hardware still works but I think the compact flash
card died a couple of years ago IIRC.  Under the hood it was just a
standard SBC (single board computer, an ALIX board IIRC).  You can do this
on your own with iptables and tc (the Linux traffic control daemon).  The
problem you are going to have is that some protocols like TCP likely will
not even work with multi-minute latencies, things get pretty hairy with
just a couple of seconds.  After 2 to 3 seconds of latencies a lot of
TCP/IP stacks will just error out.  lartc.org has a lot of good resources
in working with these tools, though their focus is on reducing contention
and latency as opposed to the opposite.

I seem to remember Vint Cerf was designing a deep space protocol to
facilitate a civilian deep space network.  I am not sure what happened with
this (
https://www.google.com/search?q=vint+cerf+deep+space+network&oq=vint+cerf+deep+space+network
).

 I have been thinking of something similar for the past couple of years,
but with a terrestrial focus in areas with low bandwidth and periodic
connectivity in mind.  Stitching together MESH networking at the lower
layers, with store and forward (bittorrent like) at the application layer
with native binary support (no base64 encoding nonsense) is not an easy
thing.

I look forward to your presentation on this topic :)

I will dust off the Network Nightmare box.  It is yours if you want it.

On Sat, Jan 5, 2019 at 1:52 PM Greg King <[email protected]> wrote:

> When I was working with Riverbed WAN accelerators I had a small 2 NIC
> device that did exactly that - insert configurable network delays to show
> how the Riverbed product could accelerate/mask network traffic latency. I
> believe it was call Network Nightmare. I don't know if it went up to 20
> minutes delay tho...
>
> I gave it to Gustin
>
> Good luck.
>
> Greg
>
> *From: *"sam khangyi" <[email protected]>
> *To: *"[email protected] [email protected]" <[email protected]>
> *Sent: *Saturday, January 5, 2019 9:18:55 AM
> *Subject: *[clug-talk] network delay loop ?
>
> How can you build a firewall / router that puts a fifo delay loop on
> network traffic in both direction ... something like 2 seconds to 20
> minutes...
>
> So here is what i am thinking about: as we approach having a colony on
> mars, 20 light minutes away from earth ( worst case scenario ) i want to
> see how would internet work with a time delay like that.
>
> Sam Khangyi
> [email protected]
> <https://link.getmailspring.com/link/[email protected]/0?redirect=mailto%3Akhangyi%40shaw.ca&recipient=Y2x1Zy10YWxrQGNsdWcuY2E%3D>
> 403 926 0610
> [image: Open Tracking]
> _______________________________________________
> clug-talk mailing list
> [email protected]
> http://clug.ca/mailman/listinfo/clug-talk_clug.ca
> Mailing List Guidelines (http://clug.ca/ml_guidelines.php)
> **Please remove these lines when replying
> _______________________________________________
> clug-talk mailing list
> [email protected]
> http://clug.ca/mailman/listinfo/clug-talk_clug.ca
> Mailing List Guidelines (http://clug.ca/ml_guidelines.php)
> **Please remove these lines when replying
>

--000000000000b449c2057ebee4c9
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><div class=3D"gmail_default" style=3D"fon=
t-family:verdana,sans-serif">Heya Sam,=C2=A0I still have that Network Night=
mare box, it is sitting on a shelf in my basement.=C2=A0 The hardware still=
 works but I think the compact flash card died a couple of years ago IIRC.=
=C2=A0 Under the hood it was just a standard SBC (single board computer, an=
 ALIX board IIRC).=C2=A0 You can do this on your own with iptables and tc (=
the Linux traffic control daemon).=C2=A0 The problem you are going to have =
is that some protocols like TCP likely will not even work with multi-minute=
 latencies, things get pretty hairy with just a couple of seconds.=C2=A0 Af=
ter 2 to 3 seconds of latencies a lot of TCP/IP stacks will just error out.=
=C2=A0 <a href=3D"http://lartc.org">lartc.org</a> has a lot of good resourc=
es in working with these tools, though their focus is on reducing contentio=
n and latency as opposed to the opposite.<br><br>I seem to remember Vint Ce=
rf was designing a deep space protocol to facilitate a civilian deep space =
network.=C2=A0 I am not sure what happened with this (<a href=3D"https://ww=
w.google.com/search?q=3Dvint+cerf+deep+space+network&amp;oq=3Dvint+cerf+dee=
p+space+network">https://www.google.com/search?q=3Dvint+cerf+deep+space+net=
work&amp;oq=3Dvint+cerf+deep+space+network</a>).<br><br>=C2=A0I have been t=
hinking of something similar for the past couple of years, but with a terre=
strial focus in areas with low bandwidth and periodic connectivity in mind.=
=C2=A0 Stitching together MESH networking at the lower layers, with store a=
nd forward (bittorrent like) at the application layer with native binary su=
pport (no base64 encoding nonsense) is not an easy thing.=C2=A0 <br><br>I l=
ook forward to your presentation on this topic :)<br><br>I will dust off th=
e Network Nightmare box.=C2=A0 It is yours if you want it.=C2=A0</div></div=
></div><br><div class=3D"gmail_quote"><div dir=3D"ltr">On Sat, Jan 5, 2019 =
at 1:52 PM Greg King &lt;<a href=3D"mailto:[email protected]">[email protected]</=
a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0p=
x 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><d=
iv><div style=3D"font-family:arial,helvetica,sans-serif;font-size:12pt;colo=
r:rgb(0,0,0)"><div>When I was working with Riverbed WAN accelerators I had =
a small 2 NIC device that did exactly that - insert configurable network de=
lays to show how the Riverbed product could accelerate/mask network traffic=
 latency. I believe it was call Network Nightmare. I don&#39;t know if it w=
ent up to 20 minutes delay tho...=C2=A0 </div><div><br></div><div>I gave it=
 to Gustin</div><div><br></div><div>Good luck.</div><div><br></div><div>Gre=
g</div><div><br></div><div><b>From: </b>&quot;sam khangyi&quot; &lt;<a href=
=3D"mailto:[email protected]" target=3D"_blank">[email protected]</a>&gt;<br><b=
>To: </b>&quot;<a href=3D"mailto:[email protected]" target=3D"_blank">clug-=
[email protected]</a> <a href=3D"mailto:[email protected]" target=3D"_blank">clu=
[email protected]</a>&quot; &lt;<a href=3D"mailto:[email protected]" target=3D=
"_blank">[email protected]</a>&gt;<br><b>Sent: </b>Saturday, January 5, 201=
9 9:18:55 AM<br><b>Subject: </b>[clug-talk] network delay loop ?<br></div><=
br><div><div>How can you build a firewall / router that puts a fifo delay l=
oop on network traffic in both direction ... something like 2 seconds to 20=
 minutes...</div><br><div>So here is what i am thinking about: as we approa=
ch having a colony on mars, 20 light minutes away from earth ( worst case s=
cenario ) i want to see how would internet work with a time delay like that=
.</div><br><div><table cellspacing=3D"0" cellpadding=3D"0"><tbody><tr><td s=
tyle=3D"vertical-align:top"></td><td><div>Sam Khangyi</div><div style=3D"fo=
nt-size:0.9em;border-top:1px solid gray;min-width:250px;max-width:300px;mar=
gin-top:4px;padding-top:4px"><div><div><a href=3D"https://link.getmailsprin=
g.com/link/[email protected]=
/0?redirect=3Dmailto%3Akhangyi%40shaw.ca&amp;recipient=3DY2x1Zy10YWxrQGNsdW=
cuY2E%3D" target=3D"_blank">[email protected]</a><br></div><div><span style=
=3D"margin-right:8px"><a href=3D"tel:403%20926%200610" target=3D"_blank">40=
3 926 0610</a></span><br></div></div></div></td></tr></tbody></table></div>=
<img class=3D"gmail-m_-319222279270454781mailspring-open" alt=3D"Open Track=
ing" style=3D"border: 0px; width: 0px; height: 0px;" width=3D"0" height=3D"=
0"><br>_______________________________________________<br>clug-talk mailing=
 list<br><a href=3D"mailto:[email protected]" target=3D"_blank">clug-talk@c=
lug.ca</a><br><a href=3D"http://clug.ca/mailman/listinfo/clug-talk_clug.ca"=
 target=3D"_blank">http://clug.ca/mailman/listinfo/clug-talk_clug.ca</a><br=
>Mailing List Guidelines (<a href=3D"http://clug.ca/ml_guidelines.php" targ=
et=3D"_blank">http://clug.ca/ml_guidelines.php</a>)<br>**Please remove thes=
e lines when replying<br></div></div></div>________________________________=
_______________<br>
clug-talk mailing list<br>
<a href=3D"mailto:[email protected]" target=3D"_blank">[email protected]</a=
><br>
<a href=3D"http://clug.ca/mailman/listinfo/clug-talk_clug.ca" rel=3D"norefe=
rrer" target=3D"_blank">http://clug.ca/mailman/listinfo/clug-talk_clug.ca</=
a><br>
Mailing List Guidelines (<a href=3D"http://clug.ca/ml_guidelines.php" rel=
=3D"noreferrer" target=3D"_blank">http://clug.ca/ml_guidelines.php</a>)<br>
**Please remove these lines when replying<br>
</blockquote></div>

--000000000000b449c2057ebee4c9--


--===============3595397165983652676==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
clug-talk mailing list
[email protected]
http://clug.ca/mailman/listinfo/clug-talk_clug.ca
Mailing List Guidelines (http://clug.ca/ml_guidelines.php)
**Please remove these lines when replying

--===============3595397165983652676==--