BigZones SDB Question

Michael Lang michi+bind-users at
Mon Feb 14 16:59:30 UTC 2005

On Mon, 2005-02-14 at 17:33, Jorg B. wrote:
> I have had the same problem with the rbl-plus zone.... The solution I 
> came up with was to run rbldnsd with a perl script that I found on the web.
> Get rbldnsd from
> and the perl scripts to convert the RBL into a 
> format that rbldnsd can understand from
> Running rbldnsd works great... Currently the mail-abuse RBL uses only 
> about 34 MB of RAM using rbldsnd !

yes but that is, what i didnt wanted to do, or i think that its the
ugliest solution to fix it.

one better solution would be to do an IXFR on a SOA change an add the
delta the database. But i coulnd figure out howto do an IXFR Query with 
Net::DNS or pythons dnspython. Ive read RFC 1995 and tryed it with
dnspython by setting 
	Flag = AA 
	Opcode = 0 # = QUERY
	added to the AuthorityList a SOA Object # queryed before 
but dont get the response im waiting for.  
Log on the Server says:
named.query:Feb 14 16:45:40.780 client query: IN  IXFR
Feb 14 17:04:05.870 client transfer of
'': IXFR poll up to date

Response on the Client says:
id 8308
opcode QUERY
flags QR RA
;ANSWER 86400 IN SOA 3 10800 3600 604800 86400

can anyone point me to what im doig wrong on the IXFR Query ? 
Kind regards
Michael Lang

Joerg, thanks anyway :)

> JB
> Michael Lang wrote:
> >Hi to all,
> >
> >i didnt find anything about setting a SlaveZone to a SDB Backend.
> >
> >What i want to do is, i get a big Zone (130-170M) from example:
> > These Zones i want to store in a SDB Backend
> >like pgsql, as the machine gets off when big changes or IXFR doesnt work
> >for some reason.
> >
> >Ive also tryed setting the cleaning-interval to something very low like
> >10 Minutes to avoid big changes in memory at once.
> >Did anyone get SDB for Slaves working ? or does anyone has another
> >solution ?
> >
> >Kind regards
> >Michael Lang
> >
> >
> >  
> >
Michael Lang <michi+bind-users at>

More information about the bind-users mailing list