What is the best way to move from using “deny duplicates" to "allow duplicates" ?

[Bjarne Blichfeldt]

Hi All

What will happen to the existing lease database when I move from “deny duplicates” to “allow duplicates” ?
Will I have to clear the lease database before I restart dhcpd with "allow duplicates" set or will it just work ? And how is it going to influence the failover if I
change one server at a time?


More details :

Two RHEL5 machines, isc-dhcpd-4.1-ESV-R3 in failover configuration, about 150.000 leases in file..

We are seeing  lot of issues with pxe clients, client id and so on.

Scenario   – pxe boot/install a thick client :
client pxe boots,  gets an address with lease time 3600 (mctl)
starts winpe which request an address, but with different clientid => new ip, lease time 3600
  winpe does not know how to renew its lease (known problem) and dies after 3600 seconds.

Also we see a lot of double registrations in dns in a VDI environment.


To make this work I would like to change to "allow duplicates", forcing the DHCP server to use the mac address, which does not change, as reference instead of client-id, which does change.

I remember from many discussion on this list, that this would do the trick.
Double checking the manual however, leaves me not so sure :
"Host declarations can match client messages based on the DHCP Client Identifier option or based on the client’s network hardware type and MAC address.
  If the MAC address is used, the host declaration will match any client with that MAC address - even clients with different client identifiers
"

Since "host" declarations is mentioned her, does that means it will only work for host declarations = fixed IP, or will "allow duplicate" actually
work for ALL dhcp leases ?




Thanks in advance for any input
Bjarne Blichfeldt

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20140403/f060f629/attachment.html>