 |
| ||||
2.0 Definitions
Port triggering is a configuration option on a NAT-enabled router that allows a host machine to dynamically and automatically forward a specific port back to itself. Port triggering opens an incoming port when your computer is using a specified outgoing port for specific traffic. by Bill_MI edited by Johkal Port forwarding, sometimes referred to as port triggering, is a process of configuring a router to make a computer or network device that is connected to it accessible to other computers and network devices from outside of the local network. by Johkal UPnP NAT Traversal FAQ Do you have questions about Universal Plug and Play (UPnP) and Network Address Translation (NAT)? This article provides answers to some of the most commonly asked questions about a variety of UPnP, NAT, and UPnP NAT Traversal issues. Posted: July 17, 2001 Q. What is UPnP? A. Universal Plug and Play (UPnP) is an architecture for pervasive peer-to-peer network connectivity of PCs and intelligent devices or appliances, particularly within the home. UPnP builds on Internet standards and technologies, such as TCP/IP, HTTP, and XML, to enable these devices to automatically connect with one another and work together to make networkingparticularly home networkingpossible for more people. Q. What does UPnP mean to the consumer? A. Simplicity, choice and more innovative experiences. Networking products that include Universal Plug and Play technology will "just work" when physically connected to the network. UPnP can work with essentially any networking media technology, wired or wireless. This includes, for example: Category 5 Ethernet cable, Wi-Fi or 802.11B wireless networks, IEEE 1394 ("Firewire"), phoneline networking or powerline networking. As these devices and PCs are connected with one another, it becomes easier for users to take advantage of innovative new services and applications. Q. What is the UPnP Forum? A. The Universal Plug and Play Forum is an open industry consortium that was formed in June 1999 to help define the UPnP standards to simplify the networking of intelligent devices in homes and, longer term, within enterprises. The forum is achieving this goal by defining and publishing UPnP device control protocols and service control protocols. As of early June 2001, more than 350 companies are members of the UPnP Forum. The UPnP Forum is directed by the 22-member UPnP Steering Committee. A Technical Committee, Marketing Committee and a variety of working committees, each focused on a specific device category, also are set up to carry out the organization's efforts. A list of forum members, along with information on joining, is available at the forum Web site. Q. What are the technical elements of UPnP? A. UPnP is broad in scope in that it targets home networks, proximity networks, and networks in small businesses and commercial buildings. It enables data communication between any two devices under the command of any control device on the network. UPnP is independent of any particular operating system, programming language, or physical medium. UPnP supports zero-configuration networking and automatic discovery, whereby a device can dynamically join a network, obtain an IP address, announce its name, convey its capabilities upon request, and learn about the presence and capabilities of other devices. DHCP and DNS servers are optional and will be used if available on the network. Furthermore, a device can leave a network smoothly and automatically without leaving any unwanted state behind. UPnP learns from the Internet's success and heavily leverages its components, including IP, TCP, UDP, HTTP, and XML. UPnP involves a multi-vendor collaboration for establishing standard Device Control Protocols (DCPs). Similar to the Internet, these are contracts based on wire protocols that are declarative, expressed in XML, and communicated via HTTP. Q. What is NAT? Why is it used? A. Network Address Translation is an Internet Engineering Task Force (IETF) standard used to allow multiple PCs or devices on a private network (using private address ranges such as 10.0.0.0 10.255.255.255, 192.168.0.0 192.168.255.255, 172.16.0.0 172.31.255.255) to share a single, globally routable IPv4 address. A main reason NAT is often deployed is because IPv4the current generation of the Internet - addresses are getting scarce. NAT is used in gateway devices that form the boundary between the public Internet and the private LAN. As IP packets from the private LAN traverse the gateway, NAT translates a private IP address and port number to a public IP address and port number, tracking those translations to keep individual sessions intact. Internet Connection Sharing in Microsoft® Windows® XP and Windows Me operating systems, along with many Internet gateway devices use NAT, particularly to connect to broadband networks such via DSL or cable modems. The use of NAT is increasing dramatically as more homes and small businesses network their PCs and share a connection to the Internet. Q. What is the problem with NAT? A. Put simply: NAT can "break" many of the compelling new PC and home networking experiences, such as multi-player games, real time communications, and other peer-to-peer services, that people increasingly want to use in their homes or small businesses. These applications will break if they use private address on the public Internet or simultaneous use of the same port number. Application must use a public address and for each session a unique port number. Large organizations have professional IT staff on hand to ensure their corporate applications can work with NAT, but smaller organizations and consumers do not have this luxury. UPnP NAT Traversal can automatically solve many of the problems the NAT imposes on applications, making this an ideal solution for small businesses and consumers. Q. Who came up with the NAT traversal solution? A. The NAT traversal solution is part of the work being done on the specification for the Internet Gateway Device (IGD) by the UPnP IGD Working Committee. UPnP member companies may join this committee or merely choose to monitor its progress. The chair of the committee is Prakash Iyer from Intel (prakash.iyer@intel.com). Many organizations, including Microsoft, drove this effort. Q. Are there other ways to solve the problem of NAT traversal? If so, why is using UPnP the best choice? A. Yes, there are other ways to solve this problem, but no other mechanism currently exists as an industry standard to address this problem in an automatic way for the consumer and in such a universally-applicable way for the developer. Other approaches require either manual intervention by the user or they require special development efforts by the Internet gateway device vendor and the software developer to handle the NAT traversal needs of specific applications. As a result, UPnP is uniquely able to solve this important problem. Consumer does the work. The manual intervention methods of NAT traversal require a consumer to use a browser, a graphical user interface-based tool on the PC, or a command line interface tool on the PC to change some settings on the Internet gateway device in the home. While some technical enthusiast users have little difficulty with this, many consumers do not feel comfortable doing this. Further, many consumers may not even realize that NAT traversal problems are interfering with their use of services across the Internet. The user may be attempting to play a multi-player game or engage in some other peer-to-peer service but find he or she cannot connect for some reason. This leads to troubleshooting, support calls, customer dissatisfaction, and reluctance on the part of the user to try new services or experiences in the future. Developer does the work. To avoid requiring the consumer to solve this NAT traversal problem manually, some Internet gateway device vendors have written and included application layer gateway support into their devices. This application layer gateway software is designed with specific applications in mind. In other words, the device vendor writes and tests specific code that will automatically enable one application to go through the NAT. If the application software is updated, the application layer code the device vendor wrote may have to be updated and tested again. This one-at-a-time way of chasing the NAT traversal problem is manageable for device vendors when there are only a few peer-to-peer or relevant applications to consider, but this approach does not scale well to 100s or 1000s of applications, can be very expensive to pursue, and likely requires specific knowledge of how each of these applications function. The better way to approach this problem is to have the device vendor add software or firmware to their device once to understand UPnP and have other devices and software be able to communicate with the NAT device using this same technology. UPnP is uniquely able to fulfill this role today. Q. What does the UPnP NAT traversal solution do? A. The scenarios that UPnP-enabled NAT traversal helps ensure include: Multi-player gaming Peer-to-peer connections Real time communications Remote Assistance (a feature in Windows XP) For IHVs this solution removes the need for writing and maintaining a database of Application Layer Gateways (ALGs) to traverse the NAT. This solution will be supported by both Windows XP and Direct Play, a programming resource in Windows, so software applications written to DPlay will be able to use the UPnP solution for NAT traversal automatically. The UPnP Forum's IGD spec achieves this automatic NAT traversal by providing methods for the following: Learning public IP address Enumerating existing port mappings Adding and removing port mappings Assigning lease times to mappings Q. Which vendors are implementing the UPnP NAT traversal solution? A. Currently most of the large gateway (DSL/Cable router) vendors have announced intentions to implement the UPnP NAT traversal solution in products they ship in 2001, beginning as soon as July. They include: Microsoft in WindowsXP, Linksys, D-Link, Intel, Netgear and Buffalo Technology, and Arescom. Some of these vendors have announced they will provide firmware or software upgrades to customers who already own their devices to add support for UPnP-enabled NAT traversal. Q. How does a consumer know which Internet gateway device has UPnP support? A. Consumers can check the Web site of their Internet gateway device vendor or read the label on the product packaging to see if this feature is included. Some retailers will know about this within the next few months. In the coming months, the UPnP Forum will make available a UPnP logo that vendors can include in their product packaging, marketing materials or on the products to indicate the product meets UPnP Forum's test requirements. Q. What resources are available to developers to implement this? A. There are a number of resources, from whitepapers to interoperability testing events (PlugFests). For technical papers, please go to »www.upnp.org/resources.htm[?]. For upcoming events, please see »www.upnp.org/events.htm[?]. Microsoft provides developer information for Windows XP on MSDN Online. Q. Where can I learn more? A. Ensuring Great Experiences with NAT Traversal and Universal Plug and Play in Windows XP Feedback received on this FAQ entry:
by marvinmiller$ edited by Sunny PPPoE stands for Point-to-Point Protocol over by lev edited by Johkal Stands for Sateful Packet Inspection more info can be found in this post. by Lanik In computer networks, a DMZ (demilitarized zone) is a computer host or small network inserted as a "neutral zone" between a company's private network and the outside public network. It prevents outside users from getting direct access to a server that has company data. by Lanik edited by Johkal See this post for detailed calculations: »[general] dBm vs. milliwatt confusion by Brano edited by Lanik A hub is a device that connects PCs together. In general, what is called a hub in todays market is a "dumb" device. In a hub, when one PC sends data onto the wire, the hub simply forwards the packets to all the other devices connected to it. Each device is responsible for determining which packets are destined for it and ignoring the others. Current "hubs" typically share bandwidth between all the ports. In the days of coaxial networking, hubs were often called "bridges". Because they forward every packet that they receive, they do nothing to streamline the traffic on your local network. A switch is a little smarter than a hub, in that it records the IP and MAC addresses in a table of all the devices connected to it. Thus, when a packet is put onto the wire by one device, the switch reads the destination address information to determine if the destination device is connected to it. If it is, the switch forwards the packet ONLY to the destination device, sparing the other devices connected to it from having to read and deal with the traffic (making your network more efficient). If the switch does not recognize the destination device, then the switch sends the packet to everything connected to it, thereby requiring the devices to decide for themselves whether or not the packet is for them. In general, switches provide each device connected to them with dedicated bandwidth. A router is the "smartest" device of them all. A router records the address information of everything connected to it like a switch. But it also records the address of the next closest router in the network. (You can program this as the "default gateway.") A router reads even more of the information in the address of a packet and makes an intelligent decision about what to do with the data based on the address. For example, if a router receives an outbound packet that has a destination address that is not in it's table, it forwards the packet to the default gateway, rather than every device attached like a switch does. This is how data moves onto, and through, the Internet. Routers are also capable of looking at the source address of a data packet and making decisions based on that as well. This means they can tell the difference between traffic that originates on your network and traffic that comes from outside. Switches and hubs can't do that (at least in a home user's price range). This means that if a router receives an inbound packet that is addressed to something not attached to it, it simply drops it and your local network doesn't have to deal with it. A switch would forward it to all your networked devices and force them to decide whether or not is should be read. This can clog up your local network with useless traffic. This is also the fundamental difference between the devices and why the router is better for your application. Let's look at security for a minute. Say I'm a hacker and I get the IP of one of your computers somehow. So I send data to you. A switch will look at the destination address, recognize it, and send the packet right on to your computer. A router on the other hand, can be programmed to look at the source address as well. You could set a rule that says if a packet originates from outside your local network, do not forward it no matter what (although this would be stupid because you would never get any data -- but it could be done). Thus, a router can protect you from attacks in ways that no hub/switch ever could. This is a pretty simple view of the differences. Remember the names are not fixed in stone. There are so-called "intelligent hubs" that act as switches, and "Layer 3" switches that can do things like a router. Feedback received on this FAQ entry:
by Lanik edited by Johkal You're not alone. Here's some general help... Make it easier by observing the "Link" or Lnk" lights on most every NIC, hub, switch, router, modem, etc. These light up (on both ends) when the connection is complete and correct. Straight cables are most common. They connect UPLINK-to-DOWNLINK connectors. Period. Cross-over cables can be handy. They connect either UPLINK-to-UPLINK or DOWNLINK-to-DOWNLINK connectors. UPLINK connectors are primarily PC network cards (NICs) and WAN ports of routers. DOWNLINK connectors are pretty much everything else unless it says otherwise. For every rule some exceptions can get you. For example, a few modems have UPLINK connectors and this causes many headaches. Most ISPs that supply these also supply Cross-over cables to connect to a PC. Shared connectors. Sometimes, an UPLINK connector on a hub, switch or router is "shared" with one of the ports. You can use that port *or* that UPLINK connector - BUT NOT BOTH. Sorry, but those Link lights may not help you correct this situation. EXAMPLE: The BEFSR41 shares "Uplink" with LAN port "1". Find out what you have! A BEFSR41 router can be an expensive (but handy) cable tester. Other routers, hubs and switches should work similarly. Here's connections that will light the "Link" lights on both ends: by Bill_MI edited by Lanik | ||||
- ISP Reviews
- Review an ISP
- Latest
- GBU
- Information
- Hardware
- FAQs
- Community
- Join
- Welcome
- Members
- For Sale
- Forums
- All Forums
- Follow
- Site
- Mobile mode
-
© DSLReports.com
Est. 1999
Saturday, 26-Mar 03:49:02
Hosted by
Liquid Networks