[Dnsmasq-discuss] Running multiple instances of dnsmasq in a fail-over configuration

[richardvoigt at gmail.com]

On Thu, Dec 9, 2010 at 12:37 PM, Tom Metro <tmetro+dnsmasq at gmail.com> wrote:
> I looked through the list archives to see what was recommended for
> running multiple instances of dnsmasq in a fail-over configuration, and
> the answer seemed to be to run ISC DHCPd instead.
>
> My preference would be to stick with dnsmasq. I'm actually less
> concerned with having redundant DHCP services than redundant DNS, but
> given that the DHCP half of dnsmasq informs the DNS half of local hosts,
> there's still a need to synchronize the leases among multiple instances.
> Seeing as the slave instance wouldn't need to respond to DHCP packets,
> it would greatly simplifies the fail-over coordination.
>
> Have you considered adding support for DHCP Failover Protocol
> (http://tools.ietf.org/html/draft-ietf-dhc-failover-12)?
>
> Has anyone rigged up a lease synchronization scheme external to dnsmasq
> using the --dhcp-script feature (just mentioned in another thread) or
> perhaps DBus?

I think it would have to be dbus, since the external lease script,
even in leasefile-ro mode, can only push lease information to dnsmasq
at startup.

>
>  -Tom
>
> --
> Tom Metro
> Venture Logic, Newton, MA, USA
> "Enterprise solutions through open source."
> Professional Profile: http://tmetro.venturelogic.com/
>
> _______________________________________________
> Dnsmasq-discuss mailing list
> Dnsmasq-discuss at lists.thekelleys.org.uk
> http://lists.thekelleys.org.uk/mailman/listinfo/dnsmasq-discuss
>