PLEASE NOTE: Communication
over the internet requires proper configuration for security considerations to prevent compromise
of sensitive data. If you have any doubts about your configuration, please contact DbNet for
guidance.
Although the local portion of the network appears quite similar to that discussed earlier, there are
some significant differences in the manner with which computers are assigned IP addresses. These
changes are required so that the router may properly direct incoming network traffic correctly to the
server.
Dynamic IP addressing is the default configuration for most computers and internet connections. This
configuration assumes the presence of a Dynamic Host Connection Protocol (DHCP) server, which is
usually a router. In this configuration, computers are assigned IP addresses when they are powered
up and connect to the network. The IP address assigned to a computer may change over time, hence the
term Dynamic.
Static IP addressing is fixed by an administrator and is constant. The static IP address
does not change over time, and will always point to the same computer or router. A static IP address
is generally required for a server, so that clients are able to find it over the network. Advanced
users will know of Dynamic DNS addressing, which can circumvent the need for a static IP, but this
application is not discussed here.
Network Address Translation (NAT) is a process in the router that allows network traffic from the
remote network to be forwarded to the local network. In networking parlance, the remote network is
referred to as the Wide Area Network (WAN), and the local network is referred to as the Local Area
Network (LAN). NAT may be performed for both incoming and outgoing network traffic. Most commonly,
router makers will refer to outgoing traffic as NAT, but will refer to incoming traffic as a Virtual
Server, Port Forwarding or Export Service. The important distinction to make here is that outgoing
traffic represents Local clients accessing the internet from inside the office, and incoming traffic
represents remote clients accessing the server from outside the office over the internet.
Public IP addresses are accessible by any computer connected to the internet. They are required to
be unique, and will be assigned by Internet Service Providers based on the pool of Public IP
addresses over which they have been given authority. These addresses refer to computers on the WAN.
Private IP address are accessible only by computers connected to the LAN in an office. They are
assigned by the users and are required to be from a set of IP addresses that have been reserved for
Private use. The two most common IP subnets used for private address are of the form 192.168.x.x and
10.x.x.x.
In the illustration above, the server has been set with the static IP address 192.168.1.2, and the
router, which can be seen as the Default Gateway, has the LAN address 192.168.1.1.
The critical router setting for the DbNet system to work properly is the forwarding of incoming
traffic to the server. As mentioned above, different router makers refer to this application using
different terms. The typical names are Virtual Server, Port Forwarding and Export Service. Most
router makers will reserve NAT as the term for outgoing traffic translation, although some will also
use NAT as the term for incoming traffic. Clues to look for when making this configuration setting
are a data entry field for the server private IP address. Some routers will use only one port location
as the port setting for both the router and the server. Other routers will accept a different port
on the router and the server. The packet protocol will also usually need to be specified, which will
be TCP. This is usually the default selection.
If you need help configuring your router, please call DbNet, and we will be happy to help you out.
One last setting for the router is the Dynamic Host Configuration Protocol (DHCP) setting for the
LAN private addresses. The router is considered to be the network master, and is given the privelege
of assigning IP addresses to other computers on the network. You should allow the router to set
the private addresses, so that clients will configure themselves when connecting to the network.
The important things to keep in mind when making this setting, is to set the starting IP address
higher than the server fixed IP address set previously. A good way to do this is to set aside a
block of lower numbered IP addresses. For example, setting the starting DHCP IP address as
192.168.1.32 will give you plenty of fixed IP addresses to work with. Some routers will ask for
a number of addresses in the block, to which you can reply with any number up to 223. Setting the
block size for 200 will work just fine.
Remote network diagram
A remote network is significantly more complex than a local network, and requires a deeper
understanding of networking principles in order to be configured properly. A router is required
to connect remote networks to the local network, and specific router, server and client configurations
are required for proper operation.
Remote networking concepts
There are three major concepts that need to be understood properly in order to be successful in remote
networking configuration. First is the difference between dynamic IP addresses and static IP
addresses. The second is the concept of Network Address Translation, referred to as NAT. The third
concept is the idea of Public and Private IP addresses.
Server Configuration
In order for the router to properly forward incoming network traffic to the server, the server must
have a private and static IP address on the LAN. The server IP address can be set from the network
properties box as shown below. The server IP address should be chosen as a low number in the subnet
without conflicting with the LAN address of the router. Convention holds that the router IP will be
set with ".1" as the last IP digit, so the server may usually be safely configured with ".2" as
the last IP digit.
Router Configuration
The router has two IP addresses. This is the nature of routers. They pass traffic from one subnet
to another. For most applications, the router will have one LAN address and one WAN address. The
LAN address is a private IP address on the local network and the WAN address is a public IP address
on the internet. As mentioned above, convention usually holds that the router will take the ".1"
IP address on the LAN. The WAN address of the router is set by the Internet Service Provider (ISP).
Most ISP's will have customer routers that are set with Dynamic IP addresses. This is done so that
the ISP may conserve its IP address pool. If you want a static IP address, the ISP will generally
charge you a premium for the service. For this discussion, we will assume that you have a static WAN
IP address assigned by your ISP.
Client configuration
For clients on the office LAN, the configuration techniques discussed in the section on Local Network
Configuration are sufficient. Remote clients will require a different configuration. For the remote
client, it is necessary to know the WAN address of the router assigned by the ISP. This is the
address you will need to set in the database section of the parameters screen of the DbNet program
on the remote client. If you have purchased a domain name assigned to your static WAN IP, you can
also use that as the setting. Once you have properly set the database parameter on a remote client,
that client will function properly both on the local network and over the internet. An example
setting is shown below, assuming that the WAN IP of the router is 1.2.3.4.
