HP E-Series Switch Configuration [1]-Spanning Tree

In this article, we will talk about the necessary settings for redundant operation of HP Procurve E-series network switches with each other. In a sense, we will build the center of the basic network topology. In basic network topologies, it is necessary to produce a topology in a tree structure by getting rid of the chain principle. In this way, the local network will work both more efficiently and more securely.

As seen in the picture below, the switch structure of the region 1 on the left side will not work logically. It will even enter the loop, that is, cyclic error in a short time. However, it would be more appropriate to establish a network with a tree topology, which we call tree structure, in the region 2 on the right side in terms of speed and efficiency.

However, before setting up a network in Tree structure, the first thing to consider when planning is which switch will be the root switch. For this reason, a priority value is entered in the switches as a fabrication.

This value is 0-61440. If this setting is not made, naturally, since all switches will be 61440, the numerically smallest of the switch MAC addresses will automatically make itself the root switch. Before starting the switch configuration, you will need to set this Priority value on the switch we plan as the root switch and on the other edge switches.

In this way, you will set yourself a root switch. In a sense, you will lay your network foundation on this switch. You can use spine type switches as root switch or you can use edge switches.

In terms of efficiency and speed, I recommend choosing a backbone switch. When you determine the root switch, the network topology we will set up will have a topology as follows.

In the picture above, you can see the -show spanning-tree command run on the console interface of a Root switch. Since this is a backbone switch, the priority value is set to 0. Also in the Root Port section it is written“This switch is root“. Let’s see the priority parameter that can take values up to 0, 1, 2 – 15 and the corresponding switch priority values as a list;

Root Priority        Switch Priority
0            –>            0
1            –>            4096
2            –>            8192
3            –>            12288
4            –>            16384
5            –>            20480
6            –>            24576
7            –>            28672
8            –>            32768
9            –>            36864
10          –>            40960
11           –>            45056
12           –>            49152
13           –>            53248
14           –>            57344
15           –>            61440

In systems using dual backbones as shown in the bottom picture, while one of the backbones is set as the root switch, the Root Priority value is set as“0” and the Root Priority value of the other is set as“1” for redundancy. In this topology, we have planned BB1 as the root switch. Let’s set the settings for BB1 and BB2 on the Console line;

bb1(config)#: spanning-tree priority 0

bb2(config)#:spanning-tree priority 1

The most important rule of the Spanning Tree model is that there is only 1 Root Switch in 1 Network. We can query whether a switch is a root switch with the following command:

bb1(config)#: show spanning-tree 

If This Switch is Root is written in the Root Port section in the lines on the screen, this switch is the root switch. If it says something like Port A3, that port is the port to the Root Switch.

Root Port(RP): It is the port selected as the closest port to the root on each switch.
Designated Port(DP): It is the port that leads to the root switch.

Due to redundant operation and Load Balancing, there are two backbone switches in the center connected to each other by trunking with “Spanning Tree Protocol”. Therefore, before explaining the spanning tree protocol, we will talk about the Trunk function.

TRUNK(Static):
It virtualizes 2 or more physical interfaces between switches and makes them work like a single interface. The maximum speed of the virtual interface created with Trunk will be equal to the sum of the speeds of the interfaces that form it. this setting must be made mutually on both switches. In the above diagram, Trunking is implemented on A1 and A2 interfaces between BB1 and BB2. To do trunking on switches, let’s write the following on the command line:

bb1(config)#: trunk A1,A2 trk1 lacp
bb2(config)#: trunk A1,A2 trk1 lacp

LACP(Dynamic): Similar to the Trunk function. The difference is that it works dynamically. This means that interfaces can work individually or virtually together. On the command line;

bb1(config)#: interface 1-2 lacp active