Switch Stack Priority

[djay_cdxi]

I have 2 switch installed model is C9300L-48P-4X. I didn't configure the switch priority and by default, both switch has a priority 1. Switch 1 is Active and Switch 2 is standby.

Switch 1 - Uplink Te1/1/1 is connected to a Cisco 4500 switch 1 (VSS)

Switch 2 - Uplink Te2/1/1 is connected to a Cisco 4500 switch 2 (VSS)

Will this cause a network loop if ever a switch has been removed in the stack? I've noticed that the uplink physical status of switch 2 which is up but upon checking on the switch configuration, status is down. There are devices also connected to the switch 2 which are up (reachable via ping) but upon checking on the switch config, status is down.

We had a network loop encountered 4 days ago which resulted to a high utilization on its interface (uplink) and the virtual link of our distribution switch (4500). 

Can you help clarifiy if this will trigger a network loop.

Thanks in advance.

[djay_cdxi]

Forgot to mention that both the Cisco 9300 are in 1 stack.

[balaji.bandi]

Cat 9300 you make a config which one to be active with priority config, Hope you have are configured uplink in port-channel ?

Follow below guide (Active election)  - make the config write config and reload and test.(when you get maintenance window)

https://www.cisco.com/c/en/us/products/collateral/switches/catalyst-9300-series-switches/white-paper-c11-741468.html

[djay_cdxi]

Uplink are not in Port-channel. They are connected to a distribution switch which has a VSS configured.

[balaji.bandi]

Uplink are not in Port-channel.

what mode of connectivity?  if you connect without any PO you see some loops (STP)

we need more information.

[djay_cdxi]

We don't configure PO on this setup since connection going to the 4500 is configured with VSS. We have no other issue with other floors on the building. The only difference is that this access switch is 9300 and the other existing are 2960.

[balaji.bandi]

This is not recommended configuration to connect to the core, with this you have STP Loops. with individual trunk config.

we suggest making a Port-channel configuration is a good approach.

with that ;

1. you do not see any loops

2. you get redundancy

3. high performance.

here is the example config :

https://www.cisco.com/c/en/us/support/docs/switches/catalyst-4000-series-switches/23408-140.html

[Kevin100]

Create etherchannel on core and this stack for uplinks and allow only the required vlans making that interfaces trunk, make spanning tree root primary for all vlan  on core and create interfaces on it , and enable routing

[paul driver]

Hello

Switch stack priority will have nothing to do to with network loops, but the way you configure and connect switches into the switch stack can.

Without looking at the config for the switch stack and the switches you attached to that stack and understanding how they were connected would be hard to tell what the root cause of the loop was.

[AdamF1]

As others have stated switch priority will not cause an issue or any type of loop. 
Based on you stating one link is down and 1 is up I’m assuming you are doing separate L2 links and STP is blocking one link. 
I would recommend doing LACP from and to your 4500x vss pair this way you can have both links up at once. 

can you share your config and the community can asssist? 

[djay_cdxi]

We have the same setup with other floors, same config on the uplinks. The only difference is that this model is 9300 and the other that are working is 2960. Currently, connection are stable but we only activated one uplink (Te1/1/1) in the 9300 going to the 4500 Distribution Switch. Sharing the configuration on the uplinks.

From Cisco 9300 Switch (Access Switch)

interface TenGigabitEthernet1/1/1
 description [CONN-to-CORESWITCH-VSS]
 switchport trunk allowed vlan 300,500,551
 switchport mode trunk
 switchport nonegotiate
 ip dhcp snooping trust
end

interface TenGigabitEthernet2/1/1 (We haven't activated yet)
 description [CONN-to-CORESWITCH-VSS]
 switchport trunk allowed vlan 300,500,551
 switchport mode trunk
 switchport nonegotiate
 ip dhcp snooping trust
end

Connected to the 4500 switch

interface TenGigabitEthernet1/1/1
 description CONN-to_Branch_SW1
 switchport trunk allowed vlan 300,500,551
 switchport mode trunk
 switchport nonegotiate
end

interface TenGigabitEthernet2/1/1 (Shutdown)
 description CONN-to_Branch_SW1
 switchport trunk allowed vlan 300,500,551
 switchport mode trunk
 switchport nonegotiate
 shutdown
end

[balaji.bandi]

This is not recommended configuration to connect to the core, with this you have STP Loops. with individual trunk config.

we suggest making a Port-channel configuration is a good approach.

with that ;

1. you do not see any loops

2. you get redundancy

3. high performance.

here is the example config :

https://www.cisco.com/c/en/us/support/docs/switches/catalyst-4000-series-switches/23408-140.html