[c-nsp] 2811 login issues

[Chris Wopat]

On Mon, Feb 8, 2010 at 11:00 AM,  <cisco-nsp-request at puck.nether.net> wrote:
> Subject: [c-nsp] 2811 login issues
> Message-ID:
>        <a62d17301002080611k66140bd9u35a7070a4ba6f81b at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> I have a 2811 that stopped accepting logins from its FastEthernet
> interface last week out of the blue. When this happened there were no
> config changes, router reboots, etc. It has a Multilink bundle
> unnumbered via that FastEthernet interface and it *does* accept logins
> from this direction. Config is simple, a default route via FA and a
> /24 via MU.
>
> A few other odd symptoms:
>
> - 'copy tftp flash' will work for about 12 seconds and then begin to timeout.
>
> - telnetting from the router to anywhere immediately gives
> "Destination unreachable; gateway or host down" without even really
> trying.
>
> What's even more strange is that everything works fine the first 5-10
> minutes after a reboot.
> It was running 12.4(15)XY1 and I was able to get it to 12.4(15)XY3 to
> see if it was a bug. It's running XY for support for its HWIC-4T1/E1.
>
> In an attempt to rule out an upstream routing problem I've added its
> default gateway (3.89) to the login ACL and it gives the same symptoms
> when connecting from there. It seems to be completely dropping packets
> vs rejecting them as it still does if you connect from an IP not on
> that ACL.
>
> 'debug ip packet' shows this when connecting via telnet or ssh:
>
> Feb  8 07:45:25.892 CDT: IP: s=10.170.8.18 (FastEthernet0/0),
> d=10.170.3.90, len 60, rcvd 2
> Feb  8 07:45:25.892 CDT: IP: s=10.170.8.18 (FastEthernet0/0),
> d=10.170.3.90, len 60, stop process pak for forus packet

Anyone have insight into this? I still have not come up with a
solution. I've also temporarily enabled CDP to confirm that things are
connected as they should be.

--Chris