For people who were following on this issue, i would like to post some updates. But before that, you can read what is the problem by going to GRE Tunnel ARP entry never times out! and GRE Tunnel ARP entry never times out! – part 2.
So after various troubleshooting. It was concluded that these log messages were nothing but cosmetic. Meaning although they are showing, they don’t change the behavior of the operation. It was stated that the bug causing this is CSCsa83049.
duplicate ip addr message seen on tunnel due to ARP entry not aged out A mGRE tunnel on a sup720 may report ARP entries that never age out.
This can happen when a mobile node sends a unicast ARP directed to the
default gateway, a.k.a sup720. These entries are not used by the switch to
make a forwarding decision therefore can be ignored.
Clear the ARP table using the “clear arp” command.
A solution to prevent the tunnel from either receiving or learning
the ARP entries is been investigated.
We have thought of using Automated “clear arp” using scripts. but, the issue is. “clear arp” will not work for cause we have bug CSCtf16300.
clear arp-cache is not working correctly Symptom:
“clear arp-cache” command is not removing the stale entries from arp table.
-Use cat6500 on 12.2(18)SXF16 or later.
-Use “clear ip arp x.x.x.x” command.
The solution is to change the IOS. But we do required the (SFX) IOS version to run the WLSM.
Anywho, Cisco said that they are investigating and planning on resolving bug CSCsa83049 By end of July, start of Aug. We shall wait till then.