How does it work the Secondary Server?

John Wobus jw354 at cornell.edu
Mon Apr 2 21:53:53 UTC 2007


Such "fallback" is in the hands of the caching servers.  The root (or 
TLD or other delegating) servers
don't even communicate with the servers they delegate to.  This isn't a 
fault, it's simply a design decision
aimed to put the function "out there", as a strategy of spreading the 
work involved, and to avoid saddling the
busy and vital delegating servers with added complexity.

If I (a client) query my ISP's caching server and this caching server, 
in turn, attempts to query an authoritative
server that happens to be down, the caching server will take note of 
the lack of a response and cache that fact.  While
this "negative caching" remains in effect, my queries and other 
clients' queries to the same caching server will be handled
using queries to alternative nameservers.  After a while, the caching 
server will give the offending name server another
try, and, if necessary, negative cache again.  This negative cache time 
interval is controlled by the caching server software
(such as BIND) and may well be configurable by the operator of that 
server.

As you pointed out, the root (delegating) servers distribute the load 
without respect to whether we label the server
"primary" or "secondary".  Even though we use these terms, all of a 
zone's nameservers are all equal in terms of
authority and delegation.

John Wobus

On Apr 2, 2007, at 5:12 PM, Ricardo Martinez wrote:

> Hello list.
> 	I have a general question about how the seconday/slave server works.  
> Please correct me if i'm wrong.
> I have two DNS bind servers, one is the primary server and the other, 
> as you could guess, is the secondary/slave server.   My question is 
> related to the way the requests are procesed in
>
> a.- Normal Situation
> b.- Failure of the Primary Server
> c.- Failure of the Secondary Server
>
> Can someone explain to me how this work, i'm a little bit confused 
> especially when a failure occurs.
> For example in normal situation the request from the root servers 
> arrives 50% to my primary server and 50% to my secondary, isn't?  If 
> so, what's the idea of a secondary server if primary goes down? it 
> will keep receiving requests even if the server is down?
>
> Hope that someone could explain to me this.
> Thanks in advance
>
> Regards,
> Ricardo.-
>
>



More information about the bind-users mailing list