This isn't a question but I didn't see anybody else post this. One thing we've noticed in our environment is that when a Gbic starts to go bad CRC errors start incrementing on a port. Currently LM doesn't monitor for CRC errors, that we could find. In order for this to work you have to go to the sub interface in config mode and turn on rmon statistics. Then you get OID 220.127.116.11.18.104.22.168.1.1.X.Y where X is for CRC errors and Y is the number you assigned as the process ID. For example on an HPE 5130
[HP5130-Ten-GigabitEthernet1/0/51]rmon statistics 1 owner LM
Gives you OID 22.214.171.124.