[Dnsmasq-discuss] High Availability: Part Deux

Eric Luehrsen ericluehrsen at hotmail.com
Wed Apr 22 04:15:30 BST 2015


Jonathan

If you have a second daemon running next to each DNSMASQ that just scans the shared lease file, it could issue a restart to DNSMASQ which would read the lease file (SIGUSR1). This blows out the resolving cache, but we have all seen other sledge hammer approaches that amazingly work. Because leases are a hash of MAC with a long enough lease time, people will return to work and connect with the same address.  Your secondary poll daemon could ignore renew times, only trigger when an all new IP was added.

A further enhancement, you could use your internal login system to write an alternate .../hosts file for "provisioned" machines that have had a successful login. If user authenticates, then there DHCP.LEASES file entry is copied to a more permanent location. This further reduces IP collisions and reduces cache blow out.


ERIC 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/attachments/20150421/0df729a7/attachment.html>


More information about the Dnsmasq-discuss mailing list