Two boxes with Coyote Linux... only one can connect to Charter Pipeline

Two boxes with Coyote Linux... only one can connect to Charter Pipeline

Post by Chris Hemmi » Thu, 02 Jan 2003 20:39:42



I am trying to set up a firewall with my Charter Pipeline connection.
I am using Coyote Linux (uses dhcpcd) to set up and am confused at the
behavior I am getting with one box...

I have an old P75 that I want to use as the firewall, and I have my
PII 750 that is my primary workstation running Win2k most of the time.
The PII with Charter Pipeline no problem with Win2k, and Coyote.

I set up both machines with identical Netgear nics, and created the
Coyote diskettes for each on a Linux box. The only difference between
the two is the local LAN card.

Both show the Netgear cards as eth1, both can communicate over the
LAN.

When I boot PII on Coyote it gets its IP from Charter within 2 seconds
of sending DHCP_DISCOVER.
I cycle the Cable Modem, then switch to the P75.
It times out waiting for a response after DHCP_DISCOVER.

I redid the Coyote disk for the PII to run dhcp, then booted it, got
it connected to Charter, then hooked the P75 internet nic (Netgear,
eth1) to the PII, booted P75 and got an IP instantly from PIIs dhcp.

I even shut them down, switched cards so the P75 nic would have the
same MAC address as was working in the PII.

Still timed out waiting for Charter's DHCP server response, even with
a 120 second timeout.

Is there something about the hardware configuration of the P75 that it
can't communicate to Charter? If so, what? I don't see any conflicts
in IRQ or IO address...

 
 
 

1. Charter@Home -> Charter Pipeline DHCP subnet problems


their Charter Pipeline service.  I've been trying very hard to get my Linux
box to pick up the new IP address correctly using DHCP, but I can't seem to

subnet.  I understand DHCP does a broadcast on a subnet to get the IP
address from the server.  Is there a way for me to control the subnet being
used for broadcast?

My previous/current IP address is:
65.3.xxx.yyy

It *should* be coming up as:
63.3.xxx.yyy

I'm using dhcpcd and I'm not getting errors, but I can't seem to be able to
get the correct IP address.  Is there a chance it has something to do with
the client or vendor id?

Any help greatly appreciated.

2. dhcpd on dual-nic Linux

3. How to connect two Linux boxes via Com ports having two wires only

4. TSET COMMAND

5. Connecting two networks with two linux boxes.

6. loss of display on exiting X

7. two routers (two wans) connecting to one lan using two nics.

8. __setfpucw disappeared from glibc

9. splitting (teeing) a pipeline into two further pipelines

10. Charter Pipeline DHCP trouble

11. charter pipeline

12. dhcp and Charter Pipeline

13. DHCP RH 7.1 and Charter Pipeline