Bind 9 is slow? Please help
Mark_Andrews at isc.org
Tue Mar 22 00:24:59 UTC 2005
> It was far from an exhaustive search, I only spent about five minutes
> on it if that, less time than it took me to write this post, but the
> BIND 9.3.1 configure script appears to default to autodetect for IPv6
> support, and bases IPv6 support on the presence of a sockaddr_in6
> So, if a platform is _capable_ of IPv6 it would seem that BIND 9.3.1
> will include IPv6 support in the compile.
> I've _not_ looked to see what the code does at runtime.
> rick jones
> The glass is neither half-empty nor half-full. The glass has a leak.
> The real question is "Can it be patched?"
> these opinions are mine, all mine; HP might not want them anyway... :)
> feel free to post, OR email to raj in cup.hp.com but NOT BOTH...
'named -4' is a runtime equivalent to 'configure --disable-ipv6'.
To the OP, I would be checking that you are running the
instance of named that you think you are. Check you logs
or query the server (dig +norec version.bind txt chaos @server).
BIND 9.3.1 and 9.2.5 both have code to fast retry on the
usual error codes returned when you don't have IPv6 configured.
They also do penalise the RTT estimates (they wern't being
penalised in 9.3.0 and 9.2.4). Either of these is sufficient
to remove the delay though the second will still show some
delay until the rtt estimate has been adjusted.
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: Mark_Andrews at isc.org
More information about the bind-users