could not listen on UDP socket: address in use

peaceofcrap2001 · 07-09-2004, 08:30 AM

Hey,
I am running (atleast trying to run) Bind 9.2.3 on a Clarkconnect 2.2 Office Edition (It is built on Red Hat 9 OS). I am also using the machine as a router and there is a DHCP server running on it. I have set up a DNS server on Fedora Core 1 and Red Hat 9 more than 5 times and I haven't encounter this problem before.

This is what I get as I restart 'named' in my '/var/log/messages'...

Code:

Jul  9 07:59:02 clarkconnect named: named startup succeeded
Jul  9 07:59:02 clarkconnect named[20448]: no IPv6 interfaces found
Jul  9 07:59:02 clarkconnect named[20448]: listening on IPv4 interface lo, 127.0.0.1#53
Jul  9 07:59:02 clarkconnect named[20448]: could not listen on UDP socket: address in use
Jul  9 07:59:02 clarkconnect named[20448]: creating IPv4 interface lo failed; interface ignored
Jul  9 07:59:02 clarkconnect named[20448]: listening on IPv4 interface eth0, 150.208.109.95#53
Jul  9 07:59:02 clarkconnect named[20448]: could not listen on UDP socket: address in use
Jul  9 07:59:02 clarkconnect named[20448]: creating IPv4 interface eth0 failed; interface ignored
Jul  9 07:59:02 clarkconnect named[20448]: listening on IPv4 interface eth1, 192.168.1.1#53
Jul  9 07:59:02 clarkconnect named[20448]: could not listen on UDP socket: address in use
Jul  9 07:59:02 clarkconnect named[20448]: creating IPv4 interface eth1 failed; interface ignored
Jul  9 07:59:02 clarkconnect named[20448]: listening on IPv4 interface ipsec0, 150.208.109.95#53
Jul  9 07:59:02 clarkconnect named[20448]: could not listen on UDP socket: address in use
Jul  9 07:59:02 clarkconnect named[20448]: creating IPv4 interface ipsec0 failed; interface ignored
Jul  9 07:59:02 clarkconnect named[20448]: not listening on any interfaces
Jul  9 07:59:02 clarkconnect named[20448]: command channel listening on 127.0.0.1#953
Jul  9 07:59:02 clarkconnect named[20448]: zone 1.168.192.in-addr.arpa/IN: loaded serial 1089294698
Jul  9 07:59:02 clarkconnect named[20448]: zone wbcoll.edu/IN: loaded serial 1089294513
Jul  9 07:59:02 clarkconnect named[20448]: running

I have read on google groups that the reason for this problem could be that I had forgotten to stop the previous named I had on my computer. However, I didn't have bind on this machine before. some said I need to update bind. So I have updated from bind 9.2.1 to Bind 9.2.2 and now to bind 9.2.3.

Finally, I suspected that the problem could be with port 53. so I downloaded nmap and run the following...

Code:

[root@clarkconnect etc]# nmap -A localhost
Interesting ports on localhost.localdomain (127.0.0.1):
(The 1651 ports scanned but not shown below are in state: closed)
PORT      STATE SERVICE            VERSION
22/tcp    open  ssh                OpenSSH 3.5p1 (protocol 1.99)
80/tcp    open  http               Apache httpd 2.0.40 ((Red Hat Linux))
81/tcp    open  http               Apache httpd 1.3.28 ((Unix) PHP/4.3.3 mod_ssl/2.8.15 OpenSSL/0.9.7d)
82/tcp    open  http               Apache httpd 1.3.28 ((Unix) PHP/4.3.3 mod_ssl/2.8.15 OpenSSL/0.9.7d)
443/tcp   open  ssl                OpenSSL
898/tcp   open  sun-manageconsole?
953/tcp   open  rndc?
1723/tcp  open  pptp?
10000/tcp open  http               Webmin httpd

Then I run the same command on a working BIND DNS server...

Code:

root@clarkconnect etc]# nmap -A 150.208.109.3
Interesting ports on 150.208.109.3:
(The 1651 ports scanned but not shown below are in state: closed)
PORT      STATE SERVICE     VERSION
22/tcp    open  ssh         OpenSSH 3.5p1 (protocol 1.99)
23/tcp    open  telnet      Linux telnetd
53/tcp    open  domain      ISC Bind 9.2.1
111/tcp   open  rpcbind     2 (rpc #100000)
139/tcp   open  netbios-ssn Samba smbd (workgroup: MYGROUP)
901/tcp   open  http        Samba SWAT administration server
1024/tcp  open  status      1 (rpc #100024)
3000/tcp  open  http        Ntop web interface 1.3.1 (platform: i686-pc-linux)
10000/tcp open  http        Webmin httpd

Anyways, does anybody have any idea how to solve this? Please...give me any constructive comments and suggestions. I appreciate your help!

Ambex

MrMud · 07-09-2004, 04:04 PM

Good trouble shooting skills, however, you missed a point.

First of all, you did the right thing by running nmap, but I'd recommend doing netstat (netstat -anp for maximum enjoyment) rather then nmap.

BIND is probably running, but it's on UDP port 53, not TCP port 53*. You did a scan for tcp. (Nmap does do UDP scans, but they can take a _long_ time, even locally, docs explain why).

*BIND does also run on TCP, which is why the other nmap of another dns server came up with tcp 53 open.

peaceofcrap2001 · 07-10-2004, 12:18 PM

MrMud,
Thanks for getting back to me.

I did run 'netstat' first, but I didn't get any useful information from it. Atleast, it was not different from the DNS server I have working. That is when I decided to run nmap.

I tried so many things since I posted the first time. Both 53* UDP and TCP ports were open. So I figured that there must have been some other type of DNS server was running on the Gateway computer. So I removed all the other DNS server entries from my 'dhcpd.conf' file except for the machine I was trying to set up as my DNS server (my router). I restarted my win 98 client and the internet started working.

So, I decided not to worry about the BIND DNS server. Now my Clarkconnect box is my router, DHCP server, and DNS server (Even thought I don't know what kind it is). I am guessing that Clarkconnect comes with its own built in and already configured DNS server.

Ambex