[Dnsmasq-discuss] negative caching

Cabrera Erwin-W30163 ecabrera at motorola.com
Thu May 18 16:41:15 BST 2006


Simon,

You were correct, the response for the invalid URL did not have any
records.  That explains the missing negative entry in the cache.

I'm still wondering why the negative response(s) from the DNS server did
not appear in the DNSMASQ log?  Tcpdump shows that a message was
received.


Thanks,
Erwin R. Cabrera

-----Original Message-----
From: dnsmasq-discuss-bounces at thekelleys.org.uk
[mailto:dnsmasq-discuss-bounces at thekelleys.org.uk] On Behalf Of Simon
Kelley
Sent: Wednesday, May 17, 2006 4:07 PM
To: dnsmasq discussion list
Subject: Re: [Dnsmasq-discuss] negative caching

Cabrera Erwin-W30163 wrote:
> 
> I am calling DNSMASQ version 2.25 without the "-N" option for turning 
> off negative caching.
>  
> For valid domain names, I see messages in the DNSMASQ log indicating 
> that a query was received, forwarded  to the designated DNS server, 
> and a reply received from the DNS server.  Also, the correct 
> information appears in the cache. The NSlookup application returns the

> correct address(es).
>  
> For invalid domain names, I see log messages indicating that a query 
> was received and forwarded, but no replies from the DNS server.  The 
> NSlookup application returns Host not found.  tcpdump indicates that a

> response was received from the DNS server.  I was expecting to see 
> negative cache entries in the cache, but no entries for the invalid 
> domain names appear in the cache.
>  
> I am doing something wrong or missing something?
>  

Use ethereal to examine the responses from the DNS server. Look to see
if there is an SOA record. If it's missing, then dnsmasq can't determine
the time-to-live for a negative cache entry and therefore can't generate
one. (The details of this are in RFC2308)

Many ISPs seem to turn off this information in their DNS servers (mine
is one). I suspect it's a way of reducing server load, but I'm not sure
if they realise that by inhibiting negative caching, they may in fact be
increasing the load in the servers.

Cheers,

Simon.


_______________________________________________
Dnsmasq-discuss mailing list
Dnsmasq-discuss at lists.thekelleys.org.uk
http://lists.thekelleys.org.uk/mailman/listinfo/dnsmasq-discuss



More information about the Dnsmasq-discuss mailing list