Problems with refresh of BIND database

[Kirsten Knüttel]

Hi,

I've got OpenVMS 7.3-1, TCP/IP 5.3 Eco2.
We made a change at the Windows DNS-Server, one of my VMS-Servers replicates with this server. Now I see for the name 2 different IP-Adresses - the old one and the new one.
When I want to refresh the BIND database I got an error:
TCPIP> set name_service/init
rndc: neither tcpip$etc:rndc.conf nor tcpip$etc:rndc.key was found. Use
'$rndc_confgen -a' to generate rndc.key or '$rndc_confgen' to generate rndc.conf

Is there an other way to make a refresh of the BIND database (or perhaps cleaning the cache?)? I must say, I can't stop TCPIP because the systems are productive.

Regards,

Kirsten

[Hoff]

Can you post the TCPIP command SHOW NAME output? (The node does have DNS running, right?)

As with most things IP, the stale entry should fall out on its own. Eventually.

If I wanted to try to force this issue (and I didn't might risking a transient DNS glitch), I'd probably stop and restart the DNS server from the TCPIP$CONFIG menu system, or more directly via @SYS$STARTUP:TCPIP$BIND_SHUTDOWN.COM and then @SYS$STARTUP:TCPIP$BIND_STARTUP.COM, as a test. I don't know that this would cure the cache, but it's worth a shot. (And I'd confirm the node is running DNS, too. Not all are, and I'm certainly not alone in having learned the need to (re)confirm the status of my own local boxes before proceding.)

There are some discussions around ECOs and versions, too, and V7.3-1 and V5.3 are likely off support by now. Mayhap V7.3-2 and V5.4?

There's another topic with similar errors:

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=1057983

If it's in production and this error isn't stomping on something, I'd probably leave it alone until the next maintenance window rolls around. Though you're in production now, so I'd definitely aslso look to get to a Prior Version Support (PVS) release like V7.3-2 and related during the next maintenance window. Or all the way to the current versions.

[Kirsten Knüttel]

Hi,

here the output of SHOW NAME:

BIND Resolver Parameters

Local domain: sto.modopaper.com

System

State: Started, Enabled

Transport: UDP
Domain: STO.MODOPAPER.COM
Retry: 4
Timeout: 4
Servers: LOCALHOST
Path: No values defined

Process

State: Enabled

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

I've tried on a test machine the shutdown and restart of the BIND, but nothing happens. The Problem is that we will make many changes in the near future (all printers will receive a new IP-adress...).
I know that the versions of the VMS and the TCPIP are too old, but I can't do anything (the customer doesn't want to pay for an update.....). But I think this year it will happen......

So a workaround at the moment is the local HOSTS. And for the rest of the printers I will try the IP-address direct and not the name, so that I don't need the DNS.

Regards,

Kirsten