named[1647]: sysquery: no addrs found for root NS (root.server)

[Scott Newell]

Since installing HPUX 11.0 and re-configuring DNS with the same files we were using on HPUX 10.20 we have been receiving the abover error message in our system log over and over and over. DNS seems to be working fine, just lots and lots of messages logged to the syslog. I haven't located anything in the forums or knowledge base so far indicating what is causing this problem or how to resolve it. Any help would be appreciated.

[Sachin Patel]

Hi Scott,
Check this web page it has all possible error messages for DNS

http://www.acmebw.com/askmrdns/bind-messages.htm

you need
. server file. Check your /etc/named.conf file
for example
zone "." {
type hint;
file "named.cache";
};

Here named.cache is roor server file

It should be something like
#cat /var/named/named.cache

; Initial cache data for root domain servers.
;
; <<>> DiG 2.0 <<>> @ns.internic.net . ns Last update 2/1/96

;; ANSWERS:
. 518400 NS G.ROOT-SERVERS.NET.
. 518400 NS A.ROOT-SERVERS.NET.
. 518400 NS H.ROOT-SERVERS.NET.
. 518400 NS B.ROOT-SERVERS.NET.
. 518400 NS C.ROOT-SERVERS.NET.
. 518400 NS D.ROOT-SERVERS.NET.
. 518400 NS E.ROOT-SERVERS.NET.
. 518400 NS I.ROOT-SERVERS.NET.
. 518400 NS F.ROOT-SERVERS.NET.
;
; Prep the cache (hotwire the addresses). Order does not matter.
;
;; ADDITIONAL RECORDS:
G.ROOT-SERVERS.NET. 518400 A 192.112.36.4
A.ROOT-SERVERS.NET. 518400 A 198.41.0.4
H.ROOT-SERVERS.NET. 518400 A 128.63.2.53
B.ROOT-SERVERS.NET. 518400 A 128.9.0.107
C.ROOT-SERVERS.NET. 518400 A 192.33.4.12
D.ROOT-SERVERS.NET. 518400 A 128.8.10.90
E.ROOT-SERVERS.NET. 518400 A 192.203.230.10
I.ROOT-SERVERS.NET. 518400 A 192.36.148.17
F.ROOT-SERVERS.NET. 518400 A 192.5.5.241


Sachin

[Sachin Patel]

Hi Scott,
You can get latest named.root from ftp.rs.internic.net. Login as anonymous then cd to domain directory and get the named.root file.

Sachin

[Scott Newell]

I should have posted more information...sorry.

This is DNS used for internal use only. These machines are not used for resolving internet addresses. Strictly used to resolve addresses internal to the company.

Thanks again!

[Sachin Patel]

Hi Scott,
Still you need this root server file. Just put it in place restart the named and if it does not resolve your problem you can delete that file.
I am very positive that this will solve your problem.

Sachin

[Scott Newell]

Well, I tried to use that file and it got rid of the error I was getting and gave me another one that was showing up as much or more than the prior error message:

check_root: 1 root servers after query to root server < min

The thing is your answer inadvertantly led me to the solution so thanks anyway!

I found that by changing my db.cache file to have the root.server entry as the IP address of the primary DNS server (itself), those messages are not being logged anymore. Prior to this change there were question marks in that place and worked fine that way under HPUX 10.20. Oh well, messages are not being continually logged to the syslog now, so I am happy.

Thanks again.

[Sachin Patel]

Hi Scott,
You can stop all query messages by doing

Add folloing in named.conf

logging {
category query { NULL; };
};

Stop and start named and all query will disappear.

Sachin

[sven verhaegen]

this error can be generated by some strange 'data' in the db.cache and/or db.root files called from named.boot file, do any of these files exist and are the NS records in there ok ? In db.cache for instance you should have no root server pointing to yourself , in db.root you should have one....

apparently the system complains because there is one NS record it cannot track back to the specific server so the only way of finding it out is looking at the NS records in you DB structure , starting with the db.root and db.cache is the first then the rest of the DB files , if syslog doesn't hold the IP of the system it cannot find you might try to simulate the error by going into NSLOOKUP troubleshooting mode , type nslookup and get into the nslookup menu mode then type
'set debug' and tracing is activated do some lookup test and see if you can see some replies failing , if it is in the DB files the error might not appear in this tracing however then you are back to scanning them , get out of debug mode by typing 'set no debug' then type CTRL-D to leave NSLOOKUP menu

[mw_4]

Hi Scott
I think that your problem is creating db.cache..
after upgraing..
you should create the db.cache
by typing
as below..
you can download db.cache
ftp.rs.internic.net or

#dig @a.root-server.net .ns > db.cache
You should also install dig first.
dig is installed by Compiling DNS..

Thanks.

[Sándor Szabó]

Hi Scott,

Sorry for the delayed answer...

Try to use the

options {
forward only;
};

option in you /etc/named.conf file. The forward option can be set to first or only (first is the default). If first is set, the DNS first checks the forwarders and then pursues the answer directly. For this last step - when the local server will attempt to answer the question on its own - needed the root servers. When only is set, the server will only ask the forwarders. If the forwarders don't answer, the query will fail. I hope it will solve your problem.
Cheers!

Sandor