[TriLUG] has roadrunner changed something recently?

Ryan Leathers Ryan.Leathers at globalknowledge.com
Tue Oct 12 16:07:01 EDT 2004


thanks for alerting me to the bug Mike and Steve... I should have looked
into that myself - duh


-----Original Message-----
From: Mike Broome [mailto:mbroome at employees.org]
Sent: Tuesday, October 12, 2004 3:39 PM
To: Triangle Linux Users Group discussion list
Subject: Re: [TriLUG] has roadrunner changed something recently?


Ryan,

We have a similar setup at home with a Cisco 2600 behind a TWC RR cable
modem and ran into the same issue over the weekend.  No matter what I
tried, the router would only come up with at 10.x.x.x DHCP address.  We
replaced the Cisco 2600 with a Netscreen 5GT to get things working
again.

What version of Cisco IOS are you running on the Cisco 2600?

There was a discussion at work about this saying that TWC made a change
to their DHCP servers over the weekend which triggers bug CSCdx67972 in
some version of Cisco IOS 12.1 and 12.2.  The fix for CSCdx67972 is
present in 12.2(11.8) or 12.2(11.3)T and later.  I was running a 12.2(6)
IOS image.  I haven't had a chance to upgrade to a later version to
verify that the fix for CSCdx67972 takes care of this, but I plan to do
that in the near future.

[ The details of the bug are something along the lines of the following:
the router is specifying a DHCP option that that the DHCP servers were
previously ignoring.  Now the servers are looking at that option and
think that the router is a cable modem and are giving it one of TWC's 
internal 10.x.x.x addresses.  The fix for CSCdx67972 changes the value
of that option sent in the DHCP request. ]

Mike

On Tue, Oct 12, 2004 at 02:59:31PM -0400, Ryan Leathers wrote:
> wondering if anyone else has experienced this...
> 
> Last Friday evening in the middle of an on-line frag-fest I lost my
> connection to the server.
> A few minutes later I turned up the likely problem - my router, connected
> directly to a cable modem (Motorola SB4200), could not reach much of
> anything by IP address.  Traceroutes from the router (Cisco 2600) died 6
> hops away.  I assumed at that point that there was some outage in the SP
> network so I went and watched TV.
> 
> On Saturday the problem remained so I did some more exploration.  It seems
> that the real problem has to do with DHCP.
> When I connect my Linux or Windows hosts directly to the cable modem I get
a
> lease offer which looks like this:
> 24.74.175.X /21 IP Addr
> 24.74.168.1 Gateway
> 10.42.32.1 DHCP Server
> 
> However, when I connect the router directly to the cable modem (using E0/0
> ip address dhcp) I get a lease like this:
> 10.42.44.233 /19 IP Addr
> 10.42.32.1 Gateway
> 24.25.4.112 DHCP Server
> 
> Note that with this lease I am able to reach a number of detinations on
what
> appears to be TWC's private network, but obviously I'm not reaching the
> Internet.
> 
> So what's up with this?  Is my dhcp client in the router not handshaking
> correctly anymore?  
> Insights welcomed - insidious conspiratorial allegations discouraged.
> 
> Ryan  

-- 
Mike Broome
mbroome(at)employees.org
-- 
TriLUG mailing list        : http://www.trilug.org/mailman/listinfo/trilug
TriLUG Organizational FAQ  : http://trilug.org/faq/
TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
TriLUG PGP Keyring         : http://trilug.org/~chrish/trilug.asc



More information about the TriLUG mailing list