I guess there are different ways to solve this depending on what you mean by analyzing the interface & connection.
You could just check /var/log/messages or /var/log/syslog (on a Unix System) for events that indicate that an interface came up or went down and write a search to detect the situation and assume, that if the Interface is ok the connection is ok:
Jan 25 13:10:02 host kernel: [2149765.760464] e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
Jan 25 13:10:02 host kernel: [2149765.761220] ADDRCONF(NETDEV_UP): eth0: link is not ready
Jan 25 13:10:02 host kernel: [2149765.761235] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
You could also write a script and set up a scripted input that verifies the connection ( using ping or telnet/nc or maybe wget for an http connection) and the creata a scripted input (easier) or a modular input:
Or maybe the ticketing software will write error messages if there is a connection problem and you can use that information (or you could use all 3 methods)
This app (or the windows pendant) are a good starting point there is a Dashboard called "Connection Details" that will populate if the app is configured correctly: http://splunk-base.splunk.com/apps/22314/splunk-for-unix-and-linux
-> It is also a good example to see how scripted inputs work (see the inputs.conf file and check the bin directory withi its shell scripts).