🚀 go-pugleaf

NTS is really just used to authenticate you on the network.
Gets rid of the mac address problem.
I had most of my problems with NTS for win98 (enternet 300 v1.3)
I have a Redhat box but connect through a router. The router does all the
PPPoE stuff.
Best money I ever spent.
Well almost.
Hendrix on the 4th of July was pretty good too.
R

On Thu, 6 Apr 2000 22:47:02 -0400, Rob <notfound@DAMNSPAMbellsouth.net>
wrote:
>Hendrix on the 4th of July was pretty good too.

Which concert was this?

-- 
Hal B
hburgiss@_SPAM_NOT_bellsouth.net
--

You weren't born yet

On Fri, 7 Apr 2000 08:02:33 -0400, Rob <notfound@DAMNSPAMbellsouth.net>
wrote:
>You weren't born yet

I was, and maybe even there (if it was in GA).

-- 
Hal B
hburgiss@_SPAM_NOT_bellsouth.net
--

Dave wrote:
> 
> I would like to know if anyone has had a similiar experience and what
> they have done to remedy
> the problem.

Sounds somewhat familiar.
Try checking the start-pppoe script, which initializes the ethernet
adapter to use for your connection. If it initializes it with an IP of
0.0.0.0, as my copy did.
I've found that that severely disrupts my routing functionality.

If this could be the problem, try changing the IP to something like
10.0.0.1, just so it's out of the way.

I mailed this to NTS, but didn't receive a reply to it. I assume I
either missed some docs, or it's a genuine bug. Still wrong in v1.3 of
the linux client though.

-- 
Frank van Schie

Well the readme that is on the NTS CD (in the linux directory specifically
says to
initialize the ethernet device to the ip address 0.0.0.0. I have a feeling
that this has
got to mess with the broadcast in some way...

I'll try it. I think that would make sense. Is 10.0.0.1 an in-use address
on the internet?

I would also like to get my hands on v1.3 of the NTS client. Any word on
how to get it
or do I have to call for a serviceman to bring it to my door (They do that
for me????)

Many thanks,
Dave

Frank van Schie wrote:

> Dave wrote:
> >
> > I would like to know if anyone has had a similiar experience and what
> > they have done to remedy
> > the problem.
>
> Sounds somewhat familiar.
> Try checking the start-pppoe script, which initializes the ethernet
> adapter to use for your connection. If it initializes it with an IP of
> 0.0.0.0, as my copy did.
> I've found that that severely disrupts my routing functionality.
>
> If this could be the problem, try changing the IP to something like
> 10.0.0.1, just so it's out of the way.
>
> I mailed this to NTS, but didn't receive a reply to it. I assume I
> either missed some docs, or it's a genuine bug. Still wrong in v1.3 of
> the linux client though.
>
> --
> Frank van Schie

Synaptic junction failure prevents me from providing exact location.
Georgia sounds right, I think I lived there once.
R

I found the stuff on the cd to be wrong.
Some of the scripts were screwed up as well

Dave wrote:
> 
> Well the readme that is on the NTS CD (in the linux directory specifically
> says to
> initialize the ethernet device to the ip address 0.0.0.0. I have a feeling
> that this has
> got to mess with the broadcast in some way...

The default route gets a numerical destination value of 0.0.0.0, which
is your eth interface then. Seems to me to be causing the problem.

> I'll try it. I think that would make sense. Is 10.0.0.1 an in-use address
> on the internet?

No, it's part of one of the (I think) 16 consecutive Class B networks?
Definitely free space.

> I would also like to get my hands on v1.3 of the NTS client. Any word on
> how to get it
> or do I have to call for a serviceman to bring it to my door (They do that
> for me????)

Well, personally I work at my ISP, which kinda gives me a connection
(through some people) to NTS, the people who make the stuff. You'd have
to mail/phone your provider's helpdesk to ask for whether they have also
licensed a linux client. If they did, then you can probably ask them for
it.

Another place to look at: http://www.roaringpenguin.com/
Their client seems heavily 'inspired' by the NTS Enternet software. :-)
Even v1.3 of it. So that might be a good bet otherwise. Plus it's 100%
free.

-- 
Frank

Hey thanks for the link. The default root is set up to be the ppp interface and
not the ethernet.

Let me show you my route. Its pretty common setup...
192.168.1.1     *               255.255.255.255 UH    0      0        0 eth1
255.255.255.255 *               255.255.255.255 UH    0      0        0 eth1
a.a.a.a   *               255.255.255.255 UH    0      0        0 ppp0
192.168.1.0     *               255.255.255.0   U     0      0        0 eth1
127.0.0.0       *               255.0.0.0       U     0      0        0 lo
default         a.a.a.a   0.0.0.0         UG    0      0        0 ppp0
(IP addresses have been changed to protect the innocent)
I took out all instances of eth0 since it is just a vehicle for ppp0... (Is
this good logic?)

wait a second, I think I spotted it!
when I start pppoe and the ppp0 interface is established, the ip address
in the routing table does not match the ip when I do an ifconfig... Go
figure...

Here's the ifconfig...
eth0      Link encap:Ethernet  HWaddr AA:AA:AA:AA:AA:AA
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:4762 errors:0 dropped:0 overruns:0 frame:0
          TX packets:6193 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100
          Interrupt:5 Base address:0xe400

eth1      Link encap:Ethernet  HWaddr AA:AA:AB:BB:BB:BB
          inet addr:192.168.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1125 errors:0 dropped:0 overruns:0 frame:0
          TX packets:145 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100
          Interrupt:11 Base address:0xe800

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:3924  Metric:1
          RX packets:110 errors:0 dropped:0 overruns:0 frame:0
          TX packets:110 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0

ppp0      Link encap:Point-to-Point Protocol
          inet addr:b.b.b.b  P-t-P:c.c.c.c  Mask:255.255.255.255
          UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1492  Metric:1
          RX packets:13 errors:0 dropped:0 overruns:0 frame:0
          TX packets:10 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:10
(again IP addresses have been changed to protect the innocent)

Notice a.a.a.a does not match b.b.b.b in ppp0.

Any idea what might cause this? Maybe NTS or ppp???

This is pretty screwy...

Many thanks...
Dave

Frank van Schie wrote:

> Dave wrote:
> >
> > Well the readme that is on the NTS CD (in the linux directory specifically
> > says to
> > initialize the ethernet device to the ip address 0.0.0.0. I have a feeling
> > that this has
> > got to mess with the broadcast in some way...
>
> The default route gets a numerical destination value of 0.0.0.0, which
> is your eth interface then. Seems to me to be causing the problem.
>
> > I'll try it. I think that would make sense. Is 10.0.0.1 an in-use address
> > on the internet?
>
> No, it's part of one of the (I think) 16 consecutive Class B networks?
> Definitely free space.
>
> > I would also like to get my hands on v1.3 of the NTS client. Any word on
> > how to get it
> > or do I have to call for a serviceman to bring it to my door (They do that
> > for me????)
>
> Well, personally I work at my ISP, which kinda gives me a connection
> (through some people) to NTS, the people who make the stuff. You'd have
> to mail/phone your provider's helpdesk to ask for whether they have also
> licensed a linux client. If they did, then you can probably ask them for
> it.
>
> Another place to look at: http://www.roaringpenguin.com/
> Their client seems heavily 'inspired' by the NTS Enternet software. :-)
> Even v1.3 of it. So that might be a good bet otherwise. Plus it's 100%
> free.
>
> --
> Frank

Dave wrote:
> 
> Notice a.a.a.a does not match b.b.b.b in ppp0.
> 
> Any idea what might cause this? Maybe NTS or ppp???
> 
> This is pretty screwy...

a.a.a.a appears to be your gateway. I say this because first there's a
static route to a.a.a.a through ppp0, then a default route through ppp0
with a.a.a.a as a gateway. I am similarly configured. :-)

If totally in doubt, ask your ISP about IP a.a.a.a, and ask if it could
be a gateway.

I wouldn't worry though, and it should work.
-- 
Frank

I decided to forego debugging the NTS stuff and just try the roaring
penguin stuff...

I set it up and did not have to mess with any setup and it worked for my
whole network the
first time.

Is the pppoe client shareware of freeware.

Thanks much Frank!
Dave Berry...

Frank van Schie wrote:

> Dave wrote:
> >
> > Notice a.a.a.a does not match b.b.b.b in ppp0.
> >
> > Any idea what might cause this? Maybe NTS or ppp???
> >
> > This is pretty screwy...
>
> a.a.a.a appears to be your gateway. I say this because first there's a
> static route to a.a.a.a through ppp0, then a default route through ppp0
> with a.a.a.a as a gateway. I am similarly configured. :-)
>
> If totally in doubt, ask your ISP about IP a.a.a.a, and ask if it could
> be a gateway.
>
> I wouldn't worry though, and it should work.
> --
> Frank

> > I'll try it. I think that would make sense. Is 10.0.0.1 an in-use address
> > on the internet?
>
> No, it's part of one of the (I think) 16 consecutive Class B networks?
> Definitely free space.

Actually 10.x.x.x is the reserved class A network for use in private networks.
192.168.x.x is the reserved class C.  I can't think of the 1 or 2 others at the
moment.

Randy S.

You are absolutely correct.  Not many people use this one, probably because it is so
hard to remember, ;-)

Randy S.

Frank van Schie wrote:

> "Randy S." wrote:
> >
> > Actually 10.x.x.x is the reserved class A network for use in private networks.
> > 192.168.x.x is the reserved class C.  I can't think of the 1 or 2 others at the
> > moment.
>
> I think it's 172.16.x.x - 172.31.y.y
> Which would be the 16 consecutive class B networks :-P
>
> But, as always, I could be mistaken. Isn't there an RFC on this?
> --
> Frank

"Randy S." wrote:
> 
> Actually 10.x.x.x is the reserved class A network for use in private networks.
> 192.168.x.x is the reserved class C.  I can't think of the 1 or 2 others at the
> moment.

I think it's 172.16.x.x - 172.31.y.y
Which would be the 16 consecutive class B networks :-P

But, as always, I could be mistaken. Isn't there an RFC on this?
-- 
Frank

"Randy S." wrote:
> 
> You are absolutely correct.  Not many people use this one, probably because it is so
> hard to remember, ;-)

Or because those who have a lot of hosts might want to use 10.x.x.x to
reserve space, and us home-users usually have less than 256 hosts and so
stick to 192.168.x.x (256 consecutive class C networks, which come to
think of it suits about 65k computers using a netmask of 255.255.0.0
;-))

Aah, private address space. It rules ;-p

I think I might have answer for your problem.. I have same problem here
(using Sympatico service in Toronto)

I read somewhere in the IP Masq (it is somewhere in the doc of IP Masq,
don't ask me for direct quote) that the IP Masq code that is now
distributed has problems with different MTU settings. Roughly what I
understood from it is that it has problems handling the DF (don't fragment)
packets on the lines where MTU is set to something other than 1500. Since
the PPPOE seems to set MTU to 1492, and your network will probably run on
MTU of 1500, the packets that travel with DF flag from your computer to the
network might never get out, and get dropped. I observed this behaviour
with my network (used: tcpdump -i ppp0). I just see the SYN from my
computer going out, SYN from them coming in (with DF flag) ... and....
wait.... till the timeout. Sometimes the site will indeed send some 2 or 3
kb (also with DF flag) of data before it will go into the state of waiting.
And sometimes the site will continue sending data (no DF flags) normally.

I have for now completely resigned from getting the masquerading from
working until kernel 2.4 comes out (hopefully they might fix the problem...
I am not so good C programmer to attempt this myself (besides exams are
more pressing matter). Instead I am running squid, and Leafnode for news
plus some nifty IMAP server to get all my e-mail accounts under one roof.

I hope this will help solve something..

Leszek.

Dave wrote:

> RH Linux 6.1 as my masq server. It connects to my dsl provider using a
> PPP over Ethernet
> software called NTS.  Everything from the linux box is fine... I can
> browse anything in and out
> of my home network. Now, from my Win98 machine, I can get to anything
> inside the home network.
> I can even browse about 25% of the sites I try on the internet. The
> problem is the remaining 75%.
>
> I have a feeling it is a deliberate limitation with NTS. I know that it
> is imposible to share connections
> in win 95/98/NT using the win version of NTS.
>
> I would like to know if anyone has had a similiar experience and what
> they have done to remedy
> the problem.
>
> I followed the directions for the 2.2.x kernels in the masq howto.
> <http://metalab.unc.edu/mdw/HOWTO/IP-Masquerade-HOWTO.html>
>
> Thanks,
> Dave