Hello again :)
I've gotten a bit further. After puttzing around awhile long to no
avail... I decided to grab a hub and seperate the two computers I have
been testing with and make them into their own little seperate network.
Everything worked great.
So... that leaves me needing to figure out how to get them to work with
our network that we have here.
My first question...
I can't seem to figure out if the server or if the client is haveing the
problem.
I originally thought it was the client because I had told the server to
ignore RIP packets. I had guessed that the client was some how paying
attention to the RIP packets. I then connected two clients to the
internet through server and had them more or less continuously ping the
outside internet as I did somethings... I cam to realize that one of them
lost it's connection while the other still had a connection. This also
backs up the fact that it may be the individual client that is causing a
problem.
But then the way I restore this connection is very strange... the
connection is restored when I ping either 10.10.1.1 or 192.168.1.1 from
the gateway linux server.
10.10.1.1 is a WinNT domain server (which I believe is also servering DHCP)
192.168.1.1 is a Novell Print and File server (i think, there's been a
bit of question to the accuracy of our IP table so we are not entirely sure)
I have also logged 192.168.1.1 sendinging route packets to the linux box.
Anyway, what I'm asking now, is does anyone know any way that I could
isolate this problem and see if it is server or client side?
Any other ideas that may help with this would be greatly appreciated.
Thanks a lot!
-Josh
On Thu, 3 Jun 1999, Jose M. Sanchez wrote:
> This sounds like your Novell server is exchanging RIP packets with your
> Linux box...
>
> Your Linux box "learns" the new routes after a few minutes, which are not
> valid...
> hence your connection is lost. Solution: remove RIP in Novell, you don't
> need it enabled if you
> are not using the Novell machine as a gateway... (and/or) disable the
> corresponding linux
> daemons...
>
> BTW: While I know you are using reserved IP's for both your subnets, it
> seems that you
> have different IP addresses for a single segment...
>
> Isn't this a -BIG- no-no?
>
> -JMS
>
> ----- Original Message -----
> From: <[EMAIL PROTECTED]>
> To: Audie Pierre <[EMAIL PROTECTED]>
> Cc: <[EMAIL PROTECTED]>
> Sent: Wednesday, June 02, 1999 9:59 AM
> Subject: Re: [Masq] Strange losses of connection...
>
>
> >
> > Hmmmm.... thanks for the replies...
> >
> > Well... by playing around with this somemore I can perhaps now explaina
> > litttle more specifically what is happening.
> >
> > I've got two computers I'm working with, one a linux box, acting as a
> > firewall internet gateway, and the other a Win95 machine for testing.
> >
> > I make my PPP connection (via modem with pppd and chat) once established
> > I configure my firewall/gatewaying rules.
> >
> > Everything appears to work, I can ping from both computers (the Win95 one
> > has the linux box as it's gateway) and I have web access from both, then
> > after a little while, perhaps of idleness, I try to ping something on the
> > Win95 machine... I get a "Reply from 192.168.1.1: destination host
> > unreachable"
> >
> > This is very peculier because the IP for the two machines are 10.10.1.6
> > (for the linux box) and 10.10.1.7 for the Win95 machine. We do have
> > other computers on our network with 192.168.x.x IP addresses but I can't
> > see how they would be interfering with what I'm trying to do.
> > 192.168.1.1 is a Novell Print and Fileserver.
> >
> > So for whatever reason it now does not work. So I start putzing around
> > pinging things to see if anything works. From the client pings only
> > work to the internal network. From the gateway pings work to the external
> > network. Then I try pinging the client from the gateway. Nothing
> > happens, then I try pining another maching on the local network
> > (10.10.1.1) which happens to be our DHCP server (WinNT) This ping goes
> > through, then I try 10.10.1.7 (the client) again, and it goes through,
> > then I trying pinging the internet from the client, it now works again.
> >
> > I just don't see the logic in why this is happening.
> >
> > Also sometime I can't even ping 10.10.1.1 from the gateway when this
> > happens, it seems that it doens't allow any activity through the local
> > ethernet card.
> >
> > Any ideas?
> >
> > Thanks bunches,
> > Josh
> >
> >
> > On Wed, 2 Jun 1999, Audie Pierre wrote:
> >
> > >
> > > * Isolate your DHCP server by using a static IP address on one of the
> clients.
> > > * If you suspect a client is not accessing the network:
> > > 1.) Check its IP configuration [winipcfg (95) or ipconfig (NT)]
> > > 2.) Ping your Linux gateway
> > > 3.) Check your Linux gateway's routing table (netstat -nr)
> > > 4.) Then ping the Internet from the gateway (Logically pinging the
> client(s) from
> > > the gateway should not bring the former back to life).
> > >
> > > * Another thing to consider is the PPP connection itself which might be
> timing out.
> > > However, it doesn't seem to be the case with yours because you are
> pinging right away
> > > at the Linux box ( I assume). I also assume that you are not using
> DIALD.
> > >
> > > * Are you running any dynamic routing daemon on the gateway (gated or
> routed)?
> > >
> > > I am not so sure where the problem is coming from but you could easily
> figure it out
> > > by eliminating some of the possibilities.
> > >
> > > Keep me informed on your progress.
> > >
> > >
> > > [EMAIL PROTECTED] wrote:
> > >
> > > > Hello,
> > > >
> > > > I'm running RedHat 5.2 with IP masquerading in attempt to provide
> > > > internet access to other computers in our office. As of now we do not
> > > > have our dedicated data line yet so the linux box is using a modem to
> > > > establish a ppp connection. It also has a NIC connecting it to the
> rest
> > > > of our internal network.
> > > >
> > > > The firewalling & masquerading rules have all been working for me but
> > > > then all of the sudden for no real reason nothing works anymore more.
> > > >
> > > > We are using internal ips of 10.x.x.x
> > > > Also on our network are Novell fileservers and an NT server which acts
> > > > normally as a DHCP server.
> > > >
> > > > The computer I am testing the linux gateway with is set with a static
> ip.
> > > >
> > > > The client computer seems to work fine with internet access for a
> while,
> > > > but then, perhaps after a few momements of inactivity, i'll try to do
> > > > something again and nothting will happen. So i attempt to ping an
> > > > externeal ip... that returns "network unreachable" or some other
> similar
> > > > message. I then check to make sure I can ping the gateway, that works
> > > > fine, I check to see if the gateway can ping the client, that works
> fine,
> > > > I ping a couple other things with the gateway to test connectivity and
> > > > all is well. Then I try pinging from the client again and things work
> > > > again? It seems that by pining the client from the gateway some
> > > > connection is re-established or something. I'm not quite sure what is
> > > > going on. If anyone has any insight on this I'd greatly appreciate
> it,
> > > > or if there is any more detailed information I can provide let me
> konw.
> > > > Thanks.
> > > >
> > > > Sincerely,
> > > > Josh Estelle
> >
> >
> > _______________________________________________
> > Masq maillist - [EMAIL PROTECTED]
> > http://tiffany.indyramp.com/mailman/listinfo/masq
> > Admin requests can be handled by web (above) or
> [EMAIL PROTECTED]
>
>
>
_______________________________________________
Masq maillist - [EMAIL PROTECTED]
http://tiffany.indyramp.com/mailman/listinfo/masq
Admin requests can be handled by web (above) or [EMAIL PROTECTED]
