Cisco WAN :: 3845 - Typical Cause Of High Number Of Ifdiscards?

I am seeing a high number of  discards reported on one of my monitored WAn rtr (3845). Solarwinds Orion  NPM is reporting a steady flow of error on two links, but not all links. What is the typical cause of these reported discards? I see no  error or counter increases in the >sh int xx; where xx is the  interface shown in the solarwinds reports Google tells me that "An  interface discard is simply where the device has decided to discard a  packet for some reason. The reason a device discards the packet(s) is  typically hard to tell. It could be a corrupt packet, that the device is  busy, buffer overflows, packet size issues, and several other things."
 
The Orion Network Performance Monitor pulls the discard information  reported within Orion in the RFC1213:IfTable, more specifically  IfInDiscards and IfOutDiscards. These are the OIDs for IfInDiscards and IfOutDiscards we use to gather this
 
Interface Errors and Discards:
 
  OID = ifInDiscards  1.3.6.1.2.1.2.2.1.13.
  0
  OID = ifInErrors 1.3.6.1.2.1.2.2.1.14.
  0
  OID = ifOutDiscards  1.3.6.1.2.1.2.2.1.19.

[code]....
 
there are no error and only the outgoing direction has discards? How do i debug this more or get more info on how to track down the root cause?