Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.misc -> Re: Setting up Oracle Type-4 JDBC Driver

Re: Setting up Oracle Type-4 JDBC Driver

From: Martin Smith <mfsmith_at_erols.com>
Date: 1997/07/14
Message-ID: <33CAEB10.AB23E11C@erols.com>#1/1

Next installment: I've got the Oracle Type 4 driver going, from my WIN95 machine to a remote (over the Internet) NT4 box running Oracle Workgroup 7.3 server. (Also tested with HPUX-based 7.3 server). My earlier problems were strictly failure to RTFM.

FWIW, I installed the required class files locally (on the client); I have not tried them on the server, but I assume that is no problem (famous last words, no doubt . . .) I would observe that the Oracle documentation bundled with the driver seemed to PRESUME that the class files would be on the client machine. That seemed pretty odd. Perhaps they are merely assuming that everything--client and server--is on the same machine? Whatever.

Now, on to the next issue: Here's a clip from the documentation--

"Firewall considerations

The JDBC Thin driver cannot connect to a database from behind a firewall. The firewall prevents the
browser from opening a TCP/IP socket to the database.

This problem can be solved by using a SQL*Net compliant firewall and using connect strings in the
applet that are compliant with the firewall configuration. This solution really only works for an
intranet, because the connect string is dependent on the firewall behind which the client browser is
running."

Now, I ask you: what the ^&%*&%$ use is a product that can't access a database from behind a firewall??

So the question is, is this a design f-up or a licensing ploy? Or just a beta?

Bonus question: is this related to the apparent problem connecting to an NT Oracle server via a firewall? We were told today by Oracle tech (?) support that "NT just switches ports" which annoys the firewall. Any light on this from anyone?

Come back...

Martin Smith

Bruce R. Geerdes wrote:

> On Sat, 12 Jul 1997 08:40:33 -0500, Martin Smith <mfsmith_at_erols.com>
> wrote:
> >Huh???? I thought the main point of the Type 4 was to skip SQL*Net!!
 

> >
> >What I got out of the instructions was that you have to have a TNS
> >TCP/IP Listener process going with the DBMS. I hope to get this
 tested
> >today or Monday--I'll let you know what happens...
>
> Yes, you are correct. SQLNet is not needed -- you talk directly to
> the database's listener via TCP/IP.
> --
> bgeerdes_at_igs.com
Received on Mon Jul 14 1997 - 00:00:00 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US