SharePoint 2013 Search Topology Creator

Have you built a new SharePoint 2013 search topology lately? If you have you will notice that is all done via PowerShell. There are also some rules around how we create a new topology and how we add and remove existing components. For anyone who is not ready to dive deep into PowerShell you can use the following walkthrough to auto-provision your Search topology.

First, lets take a look at the 6 core components that support SharePoint search:

• Admin
• Crawler
• Content Processing
• Analytics Processing
• Query Processing
• Index

When creating a new topology you need to create the topology either ‘brand new’ or clone from existing. Next, you add your components and/or remove existing components and finally Activate. Another thing to keep in mind is that you may have to modify your topology down the road. This script can help you do just that.

The script below is two-fold. The first part is a 2013SearchConfig.xml file that contains the configuration for your new topology. This XML configuration is below.

<SearchTopology>
    <admin>
        <servers Name="WFE1" remove="false"/>
    </admin>
    <crawler>
        <servers Name="WFE1" indexLocation="D:\SP\Searchindex" partition="0" remove="false"/>
    </crawler>
        <servers Name="WFE1" remove="false"/>
    <content>
        <servers Name="WFE1" remove="false"/>
    </content>
    <analytics>
        <servers Name="WFE1" remove="true"/>
    </analytics>
    <query>
        <servers Name="WFE1" remove="false"/>
    </query>
    <index>
        <servers Name="WFE1" remove="false"/>
    </index>
</SearchTopology>

 

Using the configuration above the new cmdlet will automagically build your topology. The features are included are:

• Adding new components to your topology on ANY server in the Farm
• Removing any component from any server in your topology
• Activate the new topology

Adding a component

If we look at the config we notice that one Admin component is being added to the first front end server called WFE1. If this server does not currently host an Admin component already, the script will automatically clone the topology, add the component and activate. No intervention is required.

If the server already has a component  of the same type installed the script will prompt you to ensure you want to add a new component as shown below.

This helps when you are looking to scale your topology.

Removing a component

Looking at the above example configuration file we notice that we also want to remove the analytics processing component off of the WFE1 server. When the script runs it is first going to create the new topology with any NEW components. It will then remove any components that you have requested. If you have told the script to remove a component you will see the prompt below:

As you can see it returns any components in that category in a 0-Based array. You select your component to remove and continue. I have entered 0 to delete the first component in the list.

NOTE: I had to run the command again and caused the order to change so my screenshot is off due the nature in which PowerShell brings in the results. I am adding in some sort abilities to keep the experience consistent.