redundant dhcp server stategy
For some time now, we have been striping our scopes over multiple dhcp servers per subnet with a fair amount of success
example:
server A, subnet 1: host range ...20 - ....150
server B, subnet 1: host range ...151 - ...254
server A, subnet 2: host range ...20 - ....150
server B, subnet 2: host range ...151 - ...254
now with isc dhcp ver 3 there is an option for dhcp failover, but it seems somewhat "awkward" & does not appear to provide much benefit over what we are doing already. Does anyone have any thoughts on this?.
Also, even though I do not have any reservations and have max-lease-time set to 7200 seconds, I get host entries that will not go away & have "never" as their expiration -WTF? Other than setting up a cron job to null out dhcp.leases, is there a way to prevent and/or get rid of these old entries?
Thanks,
Eric
|