Timeouts on forwarders??

[Cris Rhea]

I have two machines running BIND. One server is running 8.1.2 the other 
is running 9.2.0.  For non-local queries that fail, it appears as if
the 9.2.0 server takes twice as long to return "Server failed" than
the 8.1.2 server. 

In the example below, mh2000 is a client machine (Sun running Solaris 5.7
with normal Sun resolver libraries), rcf3k is running BIND 8.1.2,
and ldap is running BIND 9.2.0.

mh2000#> time nslookup 208.28.15.16 129.176.212.78 
Server:  ldap.mayo.edu
Address:  129.176.212.78

*** ldap.mayo.edu can't find 208.28.15.16: No response from server

real    1m15.02s
user    0m0.01s
sys     0m0.02s




mh2000#> time nslookup 208.28.15.16 129.176.212.214
Server:  rcf3k.mayo.edu
Address:  129.176.212.214

*** rcf3k.mayo.edu can't find 208.28.15.16: Server failed

real    0m27.73s
user    0m0.00s
sys     0m0.02s

This is a problem because our high volume email server is doing reverse
lookups that often fail (due to the normal nonsense on the Internet). Taking
twice as long to fail is causing problems.

Is there any way to tune the timeouts on the 9.2.0 server to fail
faster for forwarded queries?

I've looked at the docs and the archives, but haven't seen any timeout
issues with forwarders discussed previously (at least 8.X.X versus 9.X.X).

Thanks-

--- Cris


----
  Cristopher J. Rhea                      Mayo Foundation
  Research Computing Facility              Pavilion 2-25
  crhea at Mayo.EDU                        Rochester, MN 55905
  Fax: (507) 266-4486                     (507) 284-0587