Locating the SIP servers

Similar to e-mail servers, you will need to specify which server would serve a specific domain. The location of the SIP servers is described in RFC 3263. The first objective of location is to determine the IP, port, and transport protocol for the server based on the domain name. The second objective is to determine the address of a backup for the first proxy.

To perform these objectives, we will use a Domain Name System, more specifically, Name Authority Pointer (NAPTR) and Service (SRV) records. NAPTR records are employed to determine the transport protocol. To specify a transport protocol, you should insert the DNS records in the zone file of your DNS server. (Check the documentation of your DNS server on how to do it.) In the following code, we are enabling three protocols for this domain, TLS, TCP, and UDP. If the client supports TLS and UDP, TLS will be chosen because of the defined order in the records:

Order  pref   flags service regexp replacement
IN NAPTR   10   50    "s"   "SIPS+D2T" "" _sips._tcp.opensips.org.
IN NAPTR   20   50    "s"   "SIP+D2T"  "" _sip._tcp.opensips.org.
IN NAPTR   30   50    "s"   "SIP+D2U"  "" _sip._udp.opensips.org.

After selecting the transport protocol, it is time to select the preferred server, which is done as follows:

Service         TTL     Class P/W   Port Server
_sips._tcp.opensips.org. 86400 IN SRV 0 5 5060 sipA.opensips.org.
_sips._tcp.opensips.org. 86400 IN SRV 0 5 5060 sipB.opensips.org.
_sip._udp.opensips.org.  86400 IN SRV 0 5 5060 sipA.opensips.org.
_sip._udp.opensips.org.  86400 IN SRV 0 5 5060 sipB.opensips.org.

The terms in the preceding code are described as follows:

• Service: The symbolic name of the desired service
• TTL: The standard DNS time to live field
• Class: The standard DNS class field (this is always IN)
• Priority: The priority of the target host; a lower value means more preferred
• Weight: A relative weight for the records with the same priority
• Port: The TCP or UDP port on which the service is to be found
• Target: The canonical hostname of the machine providing the service

The configuration of the SRV records is often used to provide failovers and load sharing between the SIP servers. It is one of the easiest ways to get geographical redundancy in a SIP project.