Neophyte in a conundrum

Jared Johnson jared.johnson at tecstar.com
Tue Mar 14 18:12:06 UTC 2000


> >1. We're running Bind 8 on a Linux machine (SuSE 6.2)
We're on RHL 6.1 w/ BIND 822p5
> >2. No changes have been made to my nameserver, and none were made
> >prior to the current trouble.
Same here.
> >3. After the first incident, things seem to have cleared up on their
> >own for about a week, then went south again.
It is sparatic at these sites as well, that's why I think it's on their end.
> >4. I've restarted the named, attempted to flush any cached
> >information, and used dig to update my cache.db file.
Didn't help I know:)
> >5. I did add another nameserver on another network to my resolv.conf
> >file, which allows the server itself to resolve the address, but it
> >still won't do it for other machines on the network.
It shouldn't.  This is where the forwarders statement works.
> >6. The address we are trying to get to is behind a firewall, and does
> >seem to have port53 closed off.
> Are you referring to the address of the web server or the nameserver?
Port
> 53 is used to talk to nameservers.  Is the nameserver for the hostname in
> the URL you can't get to also behind the firewall?

> It sounds like a firewall or routing problem involving your address block.
> You need to do traceroutes in both directions between your machine and
> their nameserver.
Well, I'm in GT?'s address block and GT? configured it... so if this is true
then I'd better put a call in :)
https://www.pcbanking.washingtonmutual.com is the culprit in question.  Our
nameserver can't seem to query theirs, ifns.washingtonmutual.com if my
memory is good.  And this is the nameserver holding the afforemetioned
address.
If I use an identical machine in the same address block but outside the
firewall it works fine.  So routing is not an issue.  Firewall is number one
suspect atm.

TECSTAR Inc.,  Applied Solar Division
Information Technology Support Center
Jared H. Johnson, IS Engineer




More information about the bind-users mailing list