Tweaking STP on GS310TP

[depam]

Hi,

I am using both GS310TP and GS308T for my home lab network.  I have 2 hypervisors and shared storage that are connected to these switches configured as Adaptive Load Balancing.  The 2 ports on each devices are connected to each of these switches (GS310TP and GS308T) and switches are connected via the main router/fw.  Part of the workload I am doing is live migrating VMs from one hypervisor to another.  This works perfectly fine and I do not see any degredation on the network.  

However, I noticed that when I decided to bring down one hypervisor; the devices connected to those switches are getting disconnected for a few seconds (~45 seconds).  I cannot pinpoint exactly what causes this but I can see some errors on the switch as below:

Mar 09 2021 08:21:24: STP-6-PORT_STATE: Port GigabitEthernet8 moving from Learning to Forwarding

*Mar 09 2021 08:21:14: TRAPMGR-5-STP_TC: Bridge topology change.

*Mar 09 2021 08:21:09: STP-6-PORT_STATE: Port GigabitEthernet8 moving from Blocking to Learning

*Mar 09 2021 08:20:58: TRAPMGR-5-STP_NEW_ROOT: This bridge has changed to be the root bridge.

*Mar 09 2021 08:20:55: STP-6-PORT_STATE: Port GigabitEthernet8 moving from Forwarding to Blocking

*Mar 09 2021 08:20:55: STP-6-PORT_STATE: Port GigabitEthernet2 moving from Forwarding to Disabled

*Mar 09 2021 08:20:55: TRAPMGR-5-PORT_LINK_DOWN: Interface GigabitEthernet2 link down

The  G8 is my uplink while the G2 is the interface from my server connected to each of the switch.  My STP parameters are as of below

STP Operation Mode: RTSP

Forward BPDU while STP DIsabled: Disable

Bridge Max Age: 20

Bridge Forward Delay: 15

Spanning Tree Max Hops: 20

Can someone help to suggest what I can tweak here so that I can eliminate or at least minimize the downtime?  Thanks in advance

[schumaku]

Something in the infrastructure (not the two switches configured to use RSTP) seems to do legacy STP in this case.