Friday Tips #6, Part 1
Posted
by Chris Kawalek
on Oracle Blogs
See other posts from Oracle Blogs
or by Chris Kawalek
Published on Fri, 14 Dec 2012 21:25:57 +0000
Indexed on
2012/12/14
23:13 UTC
Read the original article
Hit count: 323
/Oracle/Oracle Virtualization
We have a two parter this week, with this post focusing on desktop virtualization and the next one on server virtualization.
Question:
Why would I use the Oracle Secure Global Desktop Secure Gateway?
Answer by Rick Butland, Principal Sales Consultant, Oracle Desktop Virtualization:
Well, for the benefit of those who might not be familiar with client connections in Oracle Secure Global Desktop (SGD), let me back up and briefly explain. An SGD client connects to an SGD server using two distinct protocols, which, by default, require two distinct TCP ports. The first is the HTTP protocol, used by the web browser to connect to the SGD webserver on TCP port 80, or if secure connections are enabled (SSL/TLS), then TCP port 443, commonly identified as the "HTTPS" port, that is, "SSL encrypted HTTP." The second protocol from the client to the server is the Adaptive Internet Protocol, or AIP, which is used for displaying applications, transferring drive mapping data, print jobs, and so on. By default, AIP uses the TCP port 3104, or port 5307 when SSL is enabled.
When SGD clients need to access SGD over a firewall, the ports that AIP requires are typically "closed"; and most administrators are reluctant, to put it mildly, to change their firewall configurations to allow AIP traffic on 3144/5307.
To avoid this problem, SGD introduced "Firewall Forwarding", a technique where, in effect, both http and AIP traffic are "multiplexed" onto a single "well-known" TCP port, that is port 443, the https port. This is also known as single-port firewall traversal. This technique takes advantage of the fact that, as a "well-known service", port 443 is usually "open", allowing (encrypted) traffic to pass. At the target SGD server, the two protocols are de-multiplexed and routed appropriately.
The Secure Gateway was developed in response to requirements from customers for SGD to support multi-stage DMZ's, and to avoid exposing SGD servers and the information they contain directly to connections from the Internet. The Secure Gateway acts as a reverse-proxy in the first-tier of the DMZ, accepting, authenticating, and terminating incoming client connections, and then re-encrypting the connections, and proxying them, routing them on to SGD servers, deeper in the network. The client no longer needs to know the name/IP address of the SGD servers in their network, they connect to the gateway, only. The gateway takes care of those internal network details.
The Secure Gateway supports the same "single-port firewall" capability as does "Firewall Forwarding", but offers the additional advantage of load-balancing incoming client connections amongst SGD array members, which could be cumbersome without a forward-deployed secure gateway. Load-balancing weights and policies can be monitored and tuned using the "Balancer Manager" application, and Apache mod_proxy_balancer directives.
Going forward, our architects recommend the use of the Secure Gateway over "Firewall Forwarding" for single-port firewall traversal, due to its architectural advantages, its greater flexibility and enhanced features.
Finally, it should be noted that the Secure Gateway is not separately priced; any licensed SGD customer may use the Secure Gateway component at no additional cost.
For more information, see the "Secure Gateway Administrator's Guide".
© Oracle Blogs or respective owner