3

While I'm testing the STP function in several different Switches, I found Some Switches use single MAC address for all LAN ports (e.g. Buffalo's BS-G2108M, Netgear's GS716Tv3), while some other Switches assign one MAC per LAN port (e.g. ELECOM's EHB-SG2A08-PL)

So why use "one MAC per port" design for STP function? Someone said that STP self-loop may not work correctly if "one MAC per port" design is not used. Is it True? And is there any other advantages or disadvantages by using "one MAC per port" design?

1 Answer 1

0

The one-MAC-per-port design decision really has nothing to do with STP since all the MAC addresses will be relatively in the same ballpark as far as value, and STP takes into account the port number when making decisions. It is likely that the separate MAC addresses per port are easier to use for the switch's STP port decision if the MAC addresses increase in value as the port number increases.

STP will not loop if the same MAC address is used for all the ports. On the contrary, it would be more likely to loop if the MAC addresses are different, but it is unlikely that such a problem would get through product testing, or survive long in implementations.

2
  • I think the STP self-loop will be happen if 2 LAN ports of one Switch are connected by a LAN cable. In this case, will STP work without one-MAC-per-port design? For current trend, which design is adopted by switch manufactures?
    – bamb00dark
    Commented Feb 17, 2016 at 6:28
  • 1
    The BPDUs will contain the bridge ID (MAC address) of the sending switch, and the switch should recognize that the MAC address is its own.
    – Ron Maupin
    Commented Feb 17, 2016 at 6:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.