Network

The LAN IP Addresses table initially contains the addresses the CommuniGate Pro software retrieved from the Server OS configuration. Correct this list to include all LAN (local networks) the CommuniGate Pro Server needs to serve.
The Network Address Lists section explains the list format.

Usually, you want all E-mail and Real-Time (VoIP/IM) clients connecting from the LAN addresses to be able to relay E-mails and Signals to any Internet destination. In this case you may want to inlcude the LAN addresses into the Client IP Addresses list.

The list of LAN IP Addresses is used to support Real-Time (voice, video, etc.) communications, so the CommuniGate Pro Server knows which addresses are belong to NAT'ed ("local") addresses, i.e. which addresses cannot be contacted directly from the Internet.

Use the Server LAN IP Address setting to select the Server own IP Address the Server OS uses to communicate with computers on the LAN.

CommuniGate Pro can provide SIP and real-time communications for remote clients located behind NAT devices, implementing the far-end NAT traversal functionality.

To detect clients located behind NATs, the Server needs to know which addresses are used on remote networks behind those NATs.
Use the WebAdmin Interface to specify the NATed Addresses. Open the Network pages in the Settings realm, then open the NATed IPs page.

If a SIP client sends a request to CommuniGate Pro and the client own network address specified in the request headers is included into the NATed IP Addresses list, while the Server has received this request from a different network address, NOT listed included into the NATed IP Addresses list, the Server decides that this client is behind a NAT.

To allow other users to make incoming calls to a client behind a NAT, CommuniGate Pro keeps the "communication hole" between the client and Server open by periodically sending dummy packets to that client.

Log Level

Use this setting to specify what kind of information the NAT Pinger component should put in the Server Log. Usually you should use the Major or Problems (non-fatal errors) levels.
The NAT Pinger component records in the System Log are marked with the NATPING tag.

Ping Clients Every

Use this setting to specify how often the Server should send its "pinging" packets.

Clients Limit

Use this setting to specify how many different NAT clients the Server can ping.

A Media Proxy is created to serve entries (users) located behind remote NAT devices.

A Media Proxy is created to relay traffic between an IPv4 and IPv6 entries.

Log

Use this setting to specify what kind of information the Proxy component should put in the Server Log. Usually you should use the Major or Problems (non-fatal errors) levels. But when you experience problems with the Proxy component, you may want to set the Log Level setting to Low-Level or All Info: in this case protocol-level or link-level details will be recorded in the System Log as well.
The Proxy component records in the System Log are marked with the UDPPROXY or the TCPPROXY tag.

UDP Ports

This setting specifies the port number range to be used for UDP proxy operations. If the CommuniGate Pro server is behind a NAT/Firewall, make sure that all UDP packets received by the NAT/Firewall for these ports are relayed to the CommuniGate Pro Server.

TCP Ports

This setting specifies the port number range to be used for TCP proxy operations. If the CommuniGate Pro server is behind a NAT/Firewall, make sure that all TCP connections received by the NAT/Firewall for these ports are relayed to the CommuniGate Pro Server.

Round-Robin Allocation

When this option is selected, UDP and TCP ports are allocated evenly using the entire port range.
When this option is not selected, UDP and TCP ports are allocated using the first (lowest) available port in the port range.

Source Port Restriction

When this option is selected, the UDP-based media from external sources is accepted when it comes from the correct IP address and port number.
When this option is not selected, only the media source IP address is checked. This may help serving certain broken devices that incorrectly specify their own media port numbers (in SDP documents).

UDP TOS Tag

Unless this option is set to OS default, the UDP-based media packets get the specified TOS (type of service) tag value. This may help you prioritize the media traffic if your network infrastructure assigned a higher priority to packets with the specified TOS tag.

Note: the FTP Module uses the ports from the TCP Ports set for Passive Mode transfers.

Log Level

Use this setting to specify what kind of information the Domain Name Resolver should put in the Server Log. Usually you should use the Major or Problems levels. In the later case you will see the information about all failed DNS lookups. If you use the RBL services, you may see a lot of failed lookups in the Log. When you experience problems with the Domain Name Resolver, you may want to set the Log Level setting to Low-Level or All Info: in this case protocol-level or link-level details will be recorded in the System Log as well.

The Resolver records in the System Log are marked with the DNR tag.

DNS Addresses

This setting specifies how the CommuniGate Pro Server selects the DNS servers to use. If the OS-specified option is selected, the Server reads the DNS server addresses from the OS. To force the server to re-read those addresses, click the Refresh buttom on the General page in the Settings section.

If the Custom option is selected, the CommuniGate Pro Server will use the DNS servers addresses listed in the text field next to this pop-up menu.

If no DNS server address is specified, the CommuniGate Pro Server uses the 127.0.0.1 address, trying to connect to a DNS server that can be running on the same computer as the CommuniGate Pro Server.

Initial Time-out

The Domain Name System uses the connectionless UDP protocol by default, and if there any network trouble, a UDP request or reposnse can be lost (while the TCP protocol automatically resends lost packets).
The Domain Name Resolver waits for a response from a DNS server for the period of time specified with this option.

If a response is not received, the Resolver resends the request, and waits twice longer, if it times out again, it can resend the request again and wait three times longer.

If you have several Domain Name Servers specified, each time the resolver needs to repeat a request, it sends it to the next DNS server in the list.

Retry Limit

This option specifies how many time the Resolver should re-send the same request if it has not received any response from a DNS server.

Note: when a request is an RBL request, the Resolver sends the same request not more than twice, and both times it uses the same (Initial) response time-out.

Concurrent Requests

This setting limits the number of concurrent requests the Resolver can send to Domain Name Servers. On a heavily-loaded Mail or Signal relay processing many thousand requests per second, this parameter should be selected after some testing: older DNS servers may crash if requested to process too many concurrent requests.

The Domain Name Resolver uses TCP connections if a DNS server sent a UDP response with the "Truncated" flag set. This feature allows the Resolver to retrieve very large records from DNS servers.

Dummy IP Addresses

This Address List setting allows you to specify network (IP) addresses that should be considered as "non-existent".

Some DNS authorities may choose to "map" all non-existant names within their domains to some special IP address(es).

When a domain name is resolved into IP addresses, the Resolver checks the first address. If this address is listed in the Dummy IP Addresses list, the Resolver returns the "unknown host/domain name" error code.

• --IPv6 NO switches the IPv6 support off, even if some local IPv6 addresses are detected.
• --IPv6 YES switches the IPv6 support on, even if no local IPv6 local address is detected, but the Server OS supports IPv6 networking.

Many CommuniGate Pro components use lists of Network (IP) Addresses. These lists are used to specify Client and Blacklisted addresses, to specify access restrictions for Listeners, etc.

This section describes the format used to specify Network Address Lists.

A Network Address List is specified as multi-line text data.

• one IP address
• an address range - two IP addresses separated with the minus (-) symbol: a range includes both IP addresses and all addresses between them
• an address and a numeric mask, separated with the slash (/) symbol.
  The mask value should be between 1 and 32 for IPv4 addresses and between 1 and 128 for IPv6 addresses. It specifes how many higher bits of the specified address are valid. The remaining lower bits of the address must be zero. The range includes all addresses with the specified higher bits.

The first IP address can be preceded with the exclamation point (!) symbol. In this case the specified address or the address range is excluded from the list composed using the preceding lines.

A comment (separated with the semicolon (;) symbol) the can be placed at the end of a line.

Lines starting with the semicolon symbol and empty lines are comment lines.

You may need to obtain a detailed Log of all communications with certain clients or remote servers.

Use the WebAdmin Interface to specify the Debug Addresses. Open the Network pages in the Settings realm, then open the Debug IPs page.

• accepts a TCP connection from an address in this list
• opens a TCP connection to an address in this list
• receives a UDP packet from an address in this list
• sends a UDP packet to an address in this list