rndc reconfig not functioning properly with multiple threads

[Jake Morrisey]

So there is no way to have one thread process all of that info and have
another thread answering queries in the meantime?
Is there any solution in Bind for this?  If Bind is not an option is there
an enterprise name server that can do this?  What about ANS from Nominum?
With Nominum, I thought ANS would only be used as the primary name server
and you would still use Bind as a secondary (which I would think you would
still have this problem when you add zones).

If anyone knows of any enterprise solution out there to add zones without
the timeouts I would be very interested.

By enterprise I mean:
1) No performance impact, the sdb backend of Bind seems to have a much lower
qps when testing with queryperf.  The same really goes with any DB backend
(powerdns, etc.).
2) Reliable and secure
3) fully RFC compliant like Bind.  Sends NOTIFY and accepts IXFR requests
for other secondary name servers.

There must be a solution out there to do this.  Just wondering what it is.

thanks,
-jm


On 10/10/06, Mark Andrews <Mark_Andrews at isc.org> wrote:
>
>
>        You have 100000 zones.  It takes time to read the config
>        file(s).  It takes time to find if the old zone exists so
>        it can be re-used or if it doesn't build a new.  To
>        {re-}establish all the acls etc. for each zone.  This is
>        all done as a single operation.
>
>