[c-nsp] Bridge devices - ARP takeover

[Graham Wooden]

Update: I could not keep the link up on the Sup32. Even hardcoding the MAC
addresses, traffic would flat stop at random times.  However, it's been up
for 4 days now on a Sup2. Specific interface/vlan config is exactly the
same. Only differences are the Sups, IOS, and linecard.

So, does anyone know what could be the issue? I really don't want them to be
hanging off of this aging hardware...

Works: 6509 - Sup2 - s222-adventerprisek9_wan-mz.122-18.SXF15a.bin -
WS-X6248-RJ-45

Doesn't:  6509 - Sup32 - s3223-advipservicesk9_wan-mz.122-33.SXI1.bin -
WS-X6348-RJ-45

On 8/14/09 7:00 AM, "Gert Doering" <gert at greenie.muc.de> wrote:

> Hi,
> 
> On Fri, Aug 14, 2009 at 06:57:08AM -0500, Graham Wooden wrote:
>> Agreed on the ip proxy-arp, but if it makes the link work for the time being
>> ... 
> 
> This would be VERY surprising - "ip proxy-arp" makes a difference only
> if one of the devices sends ARP requests for IP addresses that are
> off-link (specifically: that the router with "ip proxy-arp" knows to be
> off-link and has a route for it).
> 
> Your routers on both sides shouldn't do any ARPing for off-link addresses
> unless one of them has a static route pointing to the ethernet itself
> ("ip route 0.0.0.0 0.0.0.0 ethernet0" is a quite typical example).
> 
> dot1q-tagging the management interface sounds like a good plan, though :)
> 
> gert