HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
BladeSystem Network Blades
Showing results for 
Search instead for 
Did you mean: 

Spanning Tree Topology Change Notification on CISCO 3120x StackWise Port

Trusted Contributor

Spanning Tree Topology Change Notification on CISCO 3120x StackWise Port

Mitch had a customer experiencing Cisco 3120x issues:






I have a customer with several C7000 enclosures.  Each has two CISCO 3120X switches and four Cisco 3020g switches.  The switches are stacked together with the StackWise ports.  When the customer has a bare metal blade in the enclosure (no OS loaded yet), they get a Spanning Tree Topology Change Notification from the StackWise port when the bare metal server is powered-on.  The notification goes away once they power off the bare metal blade. 


Anyone have an idea of why this is occurring?   They’re seeing this on all of their enclosures. 




From Hongjun: 

Do they have “portfast” feature enabled on 3120/3020 downlinks facing blades? It should be design best practice to enable it when connecting to servers.


TCN generation has nothing to do with server OS. 3120/3020 has no idea which kind of os is running on blade. If you don’t have portfast configured for downlink ports, anytime you power off/on servers which causes interfaces up/down on the switch, it’ll generate TCN for SPT.


Reply from Mitch: 

Thank you for the response.  I spoke to the customer today and he confirmed that he does have portfast enabled on the downlinks and also on the uplinks going from the 3120x switches up to their core.  The customer does trunk downlinks on some of the CISCO switches in their enclosures.  He was wondering if by using trunked downlinks he may be overriding portfast. 


From Carlos: 

PortFast is not override when the uplinks are trunked. A Cisco Trunk (dot1q) means multiple VLANs going across the uplinks. This configuration does not affect anything on the downlinks ports. Now what I do find an issue is PortFast enabled on the Uplinks connected to their core. Portfast is to be enabled only on ports that connect end stations to switches. If PortFast is enabled on ports connecting two switches then Spanning-Tree loops can occur.


Here is a good Cisco document on PortFast, BPDU Filters and BPDU Guard:





Any other comments or input for Mitch??