Results 1 to 4 of 4
  1. #1
    Join Date
    Aug 2004
    Posts
    3

    Post Unanswered: Oracle Pass Thru "Gateway"

    Hi All,

    I have a technical problem (not coding) to be resolved.

    Currently there are some tedious rules in the company prohibiting the direct connection from Internet to Internal Oracle database. As the result I have to setup an Oracle server in the DMZ zone and establish a replication between the DMZ Oracle server and Internal Oracle server.

    Have to pay for licensed and connections on both Oracle servers, no good.

    Is there any known product which I can use to replace the Oracle server in the DMZ? Such as any known .NET components or Services running on the DMZ Server to pass the Oracle commands to Internal Oracle server?

    Appreciate your advice...


    Regards,
    Jack

  2. #2
    Join Date
    Sep 2003
    Location
    Virginia, USA
    Posts
    246
    You need something in the DMZ, but definitely not a copy of your database. Consider putting an app server out there, and optionally Oracle Connection Manager to help navigate through the firewalls. Oracle's Technet web site has pictures.
    MarkRem
    Author, Oracle Database 10g: From Nuts to Soup
    http://www.remidata.com/book_nuts2soup.htm

  3. #3
    Join Date
    Aug 2004
    Posts
    3

    APplication server query?

    Hi MArk,

    Thanks for your reply.

    I need to clarify what application server you mentioned here?

    We have setup an Oracle server in the DMZ-2, perhaps it's the application server you talked about?

    Maybe the team is using some connection manager, I have to check with them.

    Isn't there any other simpler product can do these tasks? (Non Oracle is fine with me really)

    Regards,
    Jack Kee

  4. #4
    Join Date
    Sep 2003
    Location
    Virginia, USA
    Posts
    246
    In an N-Tier architecture you will install the Oracle RDBMS software on one server (think of it as the backend server), then the application server software (such as Oracle 9iAS or BEA WebLogic) on another server (middle server), and your web server software (Apache, Netscape Server, etc.) on another server (front end server.) In a smaller environment such as development, you may put all of this software on the same physical server instead of three separate servers. In a medium setup you might put the web and app servers on the same box (I really don't know if Oracle 9iAS and Oracle HTTP Server should be on the same box or not, I am not an Oracle web master, just a DBA.)

    Given the above logical setup with 3 physically separate servers, put the database server behind your firewall. Packets being sent to the database must be sent to the IP address of your firewall, not your database. so, how do they eventually get to your database? Simple, products like Oracle Connection Manager.

    Back in Oracle7 before Oracle Connection Manager we used a tns config file named tnsnav.ora. (This is no longer supported, so don't use it.) This file the Oracle clients and servers the IP address of the firewall, so whenever commands or data was transmitted this file would add the firewall's IP address to the start of the packet. the packet got to the firewall and the fiewall would strip off the first address header and look for the original address header. That's how the firewall knew where to send the packets. OCM does this, and more.
    MarkRem
    Author, Oracle Database 10g: From Nuts to Soup
    http://www.remidata.com/book_nuts2soup.htm

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •