Primary DNS failover

Michael Voight mvoight at
Fri Mar 3 01:18:34 UTC 2000

Microsoft uses multimaster. (I think that's the name)
Obviously, not being standards based, you can't mix it with a UNIX
primary for the same domain. 

At least one product uses database replication. 

I think someone needs to come up with a standard and implement it.
With DDNS, it would be especially useful. Perhaps when the DHCP failover
RFC is finalized, we can get those people to work on DNS. :)

In BIND, it is rather simple to turn a secondary into a primary. 
When IXFR is working, then it is easy for the secondary to be update.
With scripting to check to see if the primary is up or down. it might be
possible to have a new primary up in seconds (under a minute).

I am in favor of having the protocol support failover.

Michael Voight
Cisco TAC

Johnny Fribert Lauridsen wrote:
> Lot of talk in this forum on this issue.  E.g. how to have multiple primaries for the
> same zone.  As we now all understand, this is not possible, per RFC, etc....
> So, how do we do it best (no more RFC and standards discussion, lets get practical now!)?
> UPS.  That is the first good thing.
> Periodic backup - That the next good thing.
> Some Cluster mechanism - Next good idea.
> Etc.
> Anyone have a good _practical_ setup for primary failover?  and the details, pls.
> Johnny

More information about the bind-users mailing list