How to update QUICKLY for clients on subdomains.

Barry Margolin barmar at alum.mit.edu
Fri Jan 13 00:13:53 UTC 2006


In article <dq50bf$12tg$1 at sf1.isc.org>,
 "Truong Tan Son" <sontt at fcv.fujitsu.com> wrote:

> Barry Margolin wrote:
> > Why?  If you want changes to propagate quickly, you don't want to keep 
> > cache a long time.
> 
> Nameserver seem lost cache after 1 hour.
> 
> Client can not query records of other subdomains after 1 hour. 

I don't understand why.  When a cached record times out, the caching 
server should go back to the authoritative server.  It sounds like you 
may have a problem with your root hints and delegation records, so it 
doesn't know how to find the authoritative server.

> 
> $TTL 3600;
> 
> 2006040102;        Serial
> 3H;                        Refresh
> 30M;                     Retry  
> 1W;                       Expire 
> 1D;                        Minimum TTL
> )
> 
> Need  I  increase   $Retry, $Refresh and $TTL ?

Retry and Refresh only affect slave servers, so they don't affect your 
situation.  Minimum TTL is used for negative caching.

> 
> 
> > Since this is all internal to your organization, I doubt that using a 
> > short TTL would overload your servers.  How many clients are they 
> > handling?
> 
> If $TTL is low, do nameservers shorten to clear cache ?

Yes, that's what TTL is, it's the amount of time that records stay in 
the cache.

> 
>  (Number clients approximate to 600 hosts.)

Then performance should not be a problem.  You could set $TTL to 60 and 
I doubt it would overload the server.

-- 
Barry Margolin, barmar at alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
*** PLEASE don't copy me on replies, I'll read them in the group ***



More information about the bind-users mailing list