How to deal with dhcpv6-snooping and replaced devices/stale bindings

[fbawb]

Hello,

I'm currently playing around with dhcp[v6]-snooping usinga limit of 1 device to somewhat limit our users in plugging in additional switches without notifying the IT department. I'm using various devices, mostly WB (2920) and WC-Series (2930F). For the most part this works as expected. Now from time to time we need to replace broken computers. Obviously the new device has a different MAC and the switch blocks the request. For legacy IP we can use clear dhcp-snooping binding port 99 to get rid of the old binding and allow the new device to fetch an IP. For IPv6 there doesn't seem to be a similar possibility to clear bindings, as such the old binding/MAC stays and we're unable to get the new device working. As far as I can tell only a reboot clears the table which isn't quite a usable solution.

Is there another way to clear v6 bindings?

Thanks

[fbawb]

For those searching along: Sadly there doesn't seem to be an equivialent so far. To avoid reboots the following workaround should help:

• Disable DHCPv6-Snooping limit for the port in question
• Deactivate/Reactivate the port to force the client to rerun DHCP
  (Only works if the endpoint is directly connected)
  
  • Re-Enable the port limit

Log should now warn about too many Clients but keep the lease active until the previous one times out.