Meraki stacking behavior || RSTP in stack ||

[Indian1]

Dear Team,

We have 5 switches in stack and each switches is connected to end users along with access point. now the question is If  switches are in stack how does STP works because I can see that some of the uplinks port are in blocking state by RSTP.

I just want to understand how does stack switches works with STP/RSTP.

Thank you for early response.

[AlexL1]

Hi Indian1,

A stack of switches is treated as a single spanning-tree node by the rest of the network. All switches in the stack share the same bridge ID, which is derived from the MAC address of the active switch (stack master).

The stack members communicate with each other via the stack ports, and the stack as a whole participates in STP/RSTP calculations.

In a stack, the active switch manages these roles and states for all member switches. If the stack master fails, a new master is elected, and the bridge ID updates to reflect the new master.

https://documentation.meraki.com/MS/Port_and_VLAN_Configuration/Spanning_Tree_Protocol_(STP)_Overvie...

When a new switch joins the stack, it adopts the bridge ID of the stack master. If the new switch has a lower ID and the root path cost is the same, it may become the stack root, triggering reconvergence.

For more detail answer, feel free to draw a simple network diagram 🙂

If you have any further questions, please don't hesitate to contact us.

If you found this post helpful, please give it kudos.
If my answer solved your problem, click "accept as solution" so that others can benefit from it 🙂

[Blue_Bird]

The outcome of the STP Election Process can be a little different when utilizing Stacking Technology. Let's follow the process for Switch 2 again below, but with the following in consideration:

• A switch stack utilizes one logical control plane
• A Meraki switch stack will utilize a generated chassis MAC address (e.g. 00:18:0A:4F:XX:XX) that is unique to a Meraki Organization
• A Meraki switch stack will converge and deterministically elect an Active Switch on boot
• Switch Stack Port IDs will increment starting with the Active switch.

1.A switch port receiving superior BPDUs. Port 1 and 2 receive the same superior BPDU from Switch 1, therefore, further steps are needed to determine the root port.

3.The port with the lowest root path cost. We can see Port 1 and 2 have equal path costs, so we still have a tie.

4. The port with the lowest sender BID. Again, Port 1 and 2 have the same BID (that of Switch 1), and therefore the final step will be used to determine the root port.

[CarolineS]

Hi @Blue_Bird - Was this response generated by AI? If so, please cite your sources, per community guidelines.

[Blue_Bird]

@CarolineS 

 I went through community guidelines and now I  Understand.

Thanks for sharing..!

[GIdenJoe]

When you have successfully created a stack you still need to port-channel/aggregate the uplink ports towards the coreswitch if that is what you mean.  And of course the core switch must also create the port-channel using LACP.  In that case you will have bonded uplinks and no blocked ports.

However if you are using different upstream switches that are not stacked (virtually or physically) or have a non-standard design then yes you will have blocked uplinks.

The behavior I have observed on MS switches is that all switches will use the common bridge-ID of the stackmaster and their portnumbers will start at a certain number depending on the switchnumber.  Port-channels get a different number alltogether.  This numbering is very important to notice in a packet capture since if you are having multiple downlinks to other switches or perhaps an MX pair that does not participate in STP.  This numbering effects what ports can be blocked by downstream devices.