DNS Resolver Issue

[Trace Trembath]

Running OpenVMS 7.2-1 and TCP/IP Services for OpenVMS V5.1 ECO5.

I have configured a system to have 2 DNS resolvers. They are configured in TCP/IP as follows:

$tcpip show name

BIND Resolver Parameters

Local domain: scada.xom.com

System

State: Started, Enabled

Transport: UDP
Domain: SCADA.XOM.COM
Retry: 1
Timeout: 1
Servers: 158.28.105.129, 158.28.105.2
Path: No values defined

Process

State: Enabled

Transport:
Domain:
Retry:
Timeout:
Servers:
Path:
$tcpip show config name

BIND Resolver Configuration

Transport: UDP
Domain: SCADA.XOM.COM
Retry: 1
Timeout: 1
Servers: 158.28.105.129, 158.28.105.2
Path: No values defined
$

The problem is that if 158.28.105.129 is not available, then the following command hangs at the BIND database section.

$tcpip show host

A show of a specific host OR a misspelled / non-existent host works fine. It's just the above wild card show host command that seems to hang for several seconds.

Is it normal for there to be a delay when the primary DNS resolver has gone away and TCP/IP has to switch to another resolver?

[Garry Fruth]

I used TCPTRACE/PORT=REMOTE=53 and did a TCPIP SHOW HOST in a similar configuration. Try it for yourself. I'm not sure of the cause of the behavior.

[Garry Fruth]

It looks like the wildcard request does 4 retries with 4 seconds between tries, regardless of the TCPIP SET NAME and TCPIP SET CONFIG NAME settings.