Managing Your Solr Index

Once you have successfully provisioned a Solr index, Websolr provides a clean, workable interface for managing it. This interface will vary slightly in appearance depending on whether or not you signed up through Websolr.com or one of our partnerships, such as Heroku. Despite these UI differences, the functionality is the same.

Here are the different overview pages you might see:

Websolr.com Dashboard:

Heroku Dashboard:

For the purposes of this documentation, subsequent screenshots will show the Websolr.com version of the UI.

Overview

The overview is the default landing page when you open up your Websolr index. It shows some key information:

• Index Name: The name of your index. All Websolr indices have a randomly-generated URL, but it helps to have a descriptive name to identify the index as well. The name can be changed at any time by clicking on the red pencil icon, changing the name to whatever you like, then clicking “Submit”.
• Index Region: This is the geographic region in which the index is running. Ideally your index is running in the same region as your application to minimize latency.
• Solr Configuration: This is the name of the Solr configuration that the index is currently using. More information about Solr configurations can be found in the “Configuration” section below.
• Index Status: This is the status of the index. Usually it will indicate PROVISIONED, but it may also indicate an error if there is a problem with the index.
• Solr Version: This is the version of Solr that the index is provisioned on.
• Index URL: This is the URL to your index. Treat this as a secret; anyone with the URL can read or write to your index, unless you have enabled Advanced Auth (see below for details).

It also shows some basic information about the index’s usage:

• Documents Used: These are the number of Solr documents in your index. Some subscription levels have limits on this metric. If you have a limit, it will be indicated here, otherwise it will just say “unlimited.”
• Data Used: This is the amount of data that your index is using. Note that Websolr indices are provisioned with two cores by default, a primary and a replica. Your data footprint is the sum of data used by each core.

Metrics

The Metrics tab shows a plethora of detailed time-series data about the performance of your index. From this tab, you can see statistics about your request latencies, connection conncurrency, request volumes, etc.

There is so much to unpack in this feature that we’ve broken out the documentation into its own article.

Request Stream

The request stream shows incoming requests in real time, including the processing time and HTTP response code. The “Top 20 Slow Requests” tab shows requests that took at least 1s to respond.

The UI does not currently display the incoming request or response body, but that is an area of active development.

Configuration

1. Click on the "Fork and customize [configuration]" link
2. This will open up a file editor.
3. Click on the file(s) you wish to edit, and make the changes needed. If your schema.xml or solrconfig.xml makes a reference to a file, it will be added to the list on the left.
4. Once you're all done making changes, give the configuration a new name and save the fork.
5. The fork will become available in the list above
6. Select the fork, then click on "Use [fork name] Configuration"
7. The index will be updated with the configuration you just created

Manage

This section allows you to upgrade or downgrade your index to a new subscription level. The menu looks like this:

There are several tiers available, each with several possible plans. A brief overview of each tier:

• Standard: these plans are provisioned on shared hardware, also known as a multitenant configuration. This allows Websolr to provide excellent economies of scale and pass the savings along to customers. This tier is ideal for small sites, startups, hobbyists, etc. They have modest limits on documents, data and daily requests.
• Business: these plans are provisioned on private hardware, also known as a single tenant configuration. While they are more expensive to operate, they also come with an uptime guarantee and a 10x service credit for downtime. This tier is ideal for large, mission critical sites. 
• Enterprise: these plans are also provisioned on private hardware, but generally require a substantial amount of resources or customization. These are ideal for very large, high-volume applications with very specific performance and privacy requirements.

More information can be found on our Pricing page, and if you have any other questions about each tier, you can also email us at support@websolr.com

To upgrade or downgrade your plan, select a tier and a capacity within that tier. Then click on the "Save" button. If you are upgrading to a paid plan and have not yet added a credit card to your account, you will see a form to add one now:

Upgrades occur instantly, in-place, and without downtime. Your URL and data will be unaffected.

Advanced Auth

Websolr features per-request authorization for requests made to your Solr index. This feature is perfect for anyone who needs to guarantee complete cyptographically secure access to their index -- or for anyone who wishes to expose their Solr URL for public read-only access while still protecting updates.

To enable Advanced Auth, click on the "Generate Token" button. This will generate a token for both read and write traffic:

The authorization scheme works by including three additional HTTP headers with each request to Solr. These headers, along with some implementation examples are detailed in a separate article, Advanced Auth.

Metadata

The Lucene Index Tool (LUKE) is "a handy development and diagnostic tool, which accesses already existing Lucene indexes and allows you to display and modify their content in several ways." In simple terms, this means that it will show you metadata and details about your index. Details like, what fields are defined and what are their settings, how many documents have been added, how many documents have been deleted, heap usage, and more.

This tool is available if the Luke request handler has been set up in your solrconfig.xml. This will have an entry like so:

<requestHandler name="/admin/luke" class="solr.admin.LukeRequestHandler"/>

Most Solr configurations will have this added by default; it would be somewhat unusual to not have it. If you don't see this entry listed in your dashboard, that means the above line is not included in your solrconfig.xml. You'll need to edit your configuration to add the requestHandler. Then the menu item will be available.

Clicking on this tab will open up the Luke details for your index. It will look something like this before data has been added:

For more information about Luke and how to use it, please consult the official documentation.

Deprovision

When you no longer need your Websolr index, you can destroy it completely by deprovisioning it. As a safety measure, you will need to enter your password to destroy the index:

When you have successfully entered your password, you will be able to click the button marked "Deprovsion":

Clicking the Deprovision button will take you back to your list of indices.