Thanks, Priscilla. I did some testing yesterday and discovered the same
two-minute timeout. however, I also have some Solaris boxes that will
be affected by this and Sasa Milic mentioned that these might retain the
old ARP entries. If so, we'll just have to manually clear the ARP
caches. It
Ooh, good idea about setting the MAC address. I'll have to check into
that. We all know it works with token ring interfaces but I'd never
played with it on an ethernet interface.
This subnet contains Windows workstations, Windows servers, Novell
servers, and Solaris servers so we'll probably
I just tried this and it appears that the behavior is a little different
on routed virtual interfaces on the 6513. I'm running 12.1(11) IOS and
with debug arp turned on I didn't see the duplicate IP test or the
gratuitous ARP.
Or, I'm just blind and I don't see it but I've run the test several
I don't think it's quite right that the routers send their ARP broadcast
replies when you simply do a shut/no shut. From the testing I have doen,
they don't even seem to do this if you physically disable and enable the
interface by removing and reinserting the cable. They only do this on a
Been awhile since I've read this list, but saw this posting and figured I'd
offer an alternative way of looking at this. I can recall a time when I had
to make a move just like this, without knowning what the mix of devices was
on that L2 network. If you don't need the original router for
Can you stand 2 minutes of downtime? What are these devices on the subnet?
If they are Windows machines, you may not have a problem. Just take 2
minutes to make your change and the entry for the default gateway will be
gone from the devices' ARP cache! The timeout for ARP entries for Windows is
Does this network contain servers? workstations? Both?
If the end-systems are running operating systems from the Northwestern
United States, you could push down a registry change involving the arp cache
timer.
If they are dhcp clients, option 35 is supposed to be associated with that
timer as
If the hosts are running snmp, and you have write access, it might be worth
seeing if you can get away with a wellfleet trick, wherein you delete
individual arp cache entries as they appear in the ipNetToMedia table (or
proprietary equivalent) by setting the ipNetToMediaType value to 2, and then
8 matches
Mail list logo
