TCPIP DNS

[Kjell Carlsson]

Prv012> tcpip sho vers

HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.6 - ECO 4
on an HP BL860c (1.59GHz/12.0MB) running OpenVMS V8.3-1H1

Prv012> tcptrace prv003
%SYSTEM-F-NOSUCHNODE, remote node is unknown
Prv012>


After update from TCPIP vers 5.6 ECO 2 to ECO 3 or ECO 4 the full domain name must be entered to start tcptrace. Even other applications suffer from this problem. Any advice?

Regards Kjell

[Volker Halle]

Kjell,

what does TCPIP SHOW NAME report ? And TCPIP SHOW HOST prv003 ?

Volker.

[Duncan Morris]

Kjell,

we are going to need a little more information about your DNS setup.

can you please post:

$ tcpip show conf name

and

$ tcpip show name

along with

$ tcpip show host prv003

Regards,

Duncan

[Kjell Carlsson]

Answer to your qustions:

Prv012> tcpip show name

BIND Resolver Parameters

Local domain: TUNNPLAT.SSAB.ORG

System

State: Started, Enabled

Transport: UDP
Domain: TUNNPLAT.SSAB.ORG
Retry: 2
Timeout: 5
Servers: bordc102, bordc103
Path: No values defined

Process

State: Enabled

Transport:
Domain:
Retry:
Timeout:
Servers:
Path:



Prv012> tcpip show conf name

BIND Resolver Configuration

Transport: UDP
Domain: TUNNPLAT.SSAB.ORG
Retry: 2
Timeout: 5
Servers: 172.21.140.13, 172.21.140.12
Path: No values defined



Prv012> tcpip show host prv003

BIND database

Server: 172.21.140.13 bordc102

Host address Host name

172.21.229.80 PRV003.TUNNPLAT.SSAB.ORG
Prv012>

Regards Kjell

[Volker Halle]

Kjell,

the config looks o.k.. I did work correctly before, did it ?

$ @SYS$MANAGER:TCPIP$DEFINE_COMMANDS

$ NSLOOKUP prv003

I see the same behaviour (at least with tcptrace) on E8.4, TCPIP T5.7. TELNET works.

Are the 'other' failing applications standard ones ?

Volker.

[Volker Halle]

Kjell,

$ TCPTRACE host

works, if host is entered in local TCPIP$HOST file.

Volker.

[Volker Halle]

Kjell,

$ DIG prv003 will probably also fail.

$ DIG prv003.TUNNPLAT.SSAB.ORG will most likely work.

The first dig will not return an address.

At least that is what I'm seeing with TCPIP T5.7.

Time for a PTR ;-)

Volker.

[Volker Halle]

Kjell,

if you look at the DIG output, you'll see that the query sent for prv003 will have a '.' added to the host name. This will cause the DNS server to fail looking up that node.

Example:

$ dig axpvms

; <<>> DiG 9.3.1 <<>> axpvms
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 24989
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;axpvms. IN A

;; Query time: 410 msec
;; SERVER: 10.20.30.40#53(10.20.30.40)
;; WHEN: Tue Nov 17 15:56:52 2009
;; MSG SIZE rcvd: 24

$ dig axpvms.invenate.local

; <<>> DiG 9.3.1 <<>> axpvms.invenate.local
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 35028
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;axpvms.invenate.local. IN A

;; ANSWER SECTION:
axpvms.invenate.local. 3600 IN A 10.20.30.203

;; Query time: 420 msec
;; SERVER: 10.20.30.40#53(10.20.30.40)
;; WHEN: Tue Nov 17 15:57:01 2009
;; MSG SIZE rcvd: 55

Volker.

[Kjell Carlsson]

Volker

Before update from ECO 2 it worked.

Nslookup works. The 'other' applications are custom applications.

Yes, I have noticed that it works if locally defined.

Right, the first $DIG doesn't work.


I also updated Alpha/OpenVMS 8.3 to ECO 4. No problem, tcptrace still works as earlier.

Regards Kjell

[Volker Halle · Accepted solution]

Kjell,


I also updated Alpha/OpenVMS 8.3 to ECO 4. No problem, tcptrace still works as earlier.


And I thought Alpha and I64 were being built from the same sources ?!

I've raised a PTR against TCPIP T5.7

Volker.