RBL & BIND 8.2.4

[Mike Diggins]

We recently began transferring the rbl-plus zone to our name servers.
Not knowing how big the zone was (20MB!) I decided to transfer it to our
single master name server using an incremental zone transfer (BIND 8.2.4).
I then transfer the zone to our primary and secondary name servers using a
standard zone transfer.

I don't really understand how IXFR works but it seems to have reduced the
number of times the zone downloads. Actually it seems to happen
consistently every few hours. However, when it transfers I get a

Jul 18 00:01:19 host1 named[28787]: [ID 295310 daemon.notice] drained
416 queries (delay 36 sec)

Does this mean my entire name server is "deaf" for the entire duration of
the transfer (36 seconds according to the message)? What can I do to
alleviate this problem? Would incremental zone transfers to my slaves
help? Can I do configure an incremental zone transfer for one zone only or
does it have to be all the zones?

Any help/advice would be appreciated.

-Mike