I just came into a situation with a few SG300 switches already in use. As part of making the networks work together, I turned on Option 82 and it does not behave as I would expect it to do. Specifically, the Remote ID is the same for everything connected to the SG300! It appears to be an identifier for the switch itself.
In my other experience (non-Cisco)...
The Circuit ID usually identifies the switch or system itself, and sometimes adds some detail about the connected port. It appears that this is how Cisco provides the Circuit ID as described in this other thread:
https://community.cisco.com/t5/small-business-switches/dhcp-option-82-sg300-how-to-decipher-circuitid-bytes/m-p/2334993/highlight/true#M10295
Further, the Remote ID is somehow unique to the port itself. This is often the port name, but could be a custom identifier. In the SG300, however, it is not unique.
I wouldn't normally ask a vendor to make a change so it would match other vendors... but in this case, the RFC documentation seems to be in conflict with how Remote IDs are implemented.
RFC3046 generally explains that the Remote ID is used to provide a trusted identifier for the remote system. When using the SG300, however, it does not provide a unique Remote ID identifier on a per port basis, so it can't be used as a trusted identifier to validate connected devices.
Specifically, RFC3046 states:
The remote ID MUST be globally unique.
Is there an option I am not noticing that allows the Remote IDs to be unique?
If not, is it possible that an option could be made to allow the Remote ID to match the "Port Description" or to match a new per-port ID?
Thank you, Chris
