Powerconnect 6248 layer 2 packet trace

hi Daniel,

I start to investigate about  a general network fault encountered some time ago.

We have 3 stacks of 6248.

From the first day all of the switches continuously raised "spanning tree topology change" events so i thought there was something wrong with STP settings.

That day network goes down i found many "new stp root" events too.

I start to investigate my network infrastructure and search info about STP protocol what it does and how it works.

I already set portfast for the end users devices and now there's no more "spanning tree topology change" every few seconds like before, so i think it's ok.

I think 've found two other issues:

STP root bridge role not on the core switch, but on a Cisco 800 series device was connected some times ago (but after the network fault day) and a link between the core switch and a peripherial unmanaged switch, blocked by STP:

#show spanning-tree blockedports

Spanning tree Enabled (BPDU flooding : Disabled) mode rstp

CST Regional Root:        80:00:5C:26:0A:DB:7D:FE

Regional Root Path Cost:  0

###### MST 0 Vlan Mapped:   1-5

ROOT ID

             Address         00:18:73:6D:9E:1D

             Path Cost       201000

             Root Port      ch4

             Hello Time 2 Sec Max Age 20 sec Forward Delay 15 sec

Bridge ID

             Priority        32768

             Address         80:00:5C:26:0A:DB:7D:FE

             Hello Time 2 Sec Max Age 20 sec Forward Delay 15 sec

Interfaces

Name    State   Prio.Nbr    Cost       Sts  Role PortFast  RestrictedPort

------  -------- --------- ----------  ---- ----- --------  -------

2/g44   Enabled  128.96       200000   DSC  Bkup      No      No

Before try new settings i would like to understand:

Are the entire network traffic passing though the root bridge?

Can i safely disconnect that disabled link on the core switch, and eventually try to force the root bridge election on itself?

On the core switch there are VLAN, tagged ports, layer 3 route etc. and i don't want to mess with them :-)

Thanks,

Mauro