[bind] Re: Suggestions for coping with this issue

Tuc at T-B-O-H.NET ml at t-b-o-h.net
Sun May 4 00:26:20 UTC 2008

> Tuc at T-B-O-H.NET wrote:
> > Hi,
> > 
> > 	We have a few systems that need to do DNS lookups for a few domains
> > outside of our control, make decisions based on it, and sleep for an hour.  
> > The problem is that they have a hair timer on getting the reply back. It seems 
> > if it isn't back in 500ms or less, it considers the domain invalid and won't 
> > allow anyone who uses that system to do what they need to.  
> > 
> > 	The domains we need to get have a 5 minute TTL on the records. After 
> > you use nslookup/dig to get the information, the subsequent requests are of 
> > course quicker due to caching. 
> > 
> > 	Is there anything we can do to "poke" bind 9 to either extend the
> > TTL, or during a special query/command flush and reget that remote zone 
> > completely?
> #!/bin/sh
> while true
>           do
>                  echo flushing
>                  rndc flushname www.cnn.com
>                  echo digging
>                  dig www.cnn.com
>                  echo sleeping
>                  sleep 2400
>                  echo next
>          done
	That seems exactly what I want.

	I went looking at rndc man page, which didn't say much. Checked
the ndc manpage too, didn't say much. I need to figure out where all the
(r)ndc commands are, I'm probably missing out on alot of possibilities 
for what I want.

	Many thanks!....

	Um, wait........

[root at ports named]# /usr/sbin/named -v
BIND 9.2.4
[root at ports named]# rndc flushname www.cnn.com
rndc: 'flushname' failed: unknown command

	Rut roh.... Is that something PAST 9.2.4?

		Thanks, Tuc

More information about the bind-users mailing list