Troubleshooting and support

The following issues present as the most common troubleshooting targets for the Content Assistant. Before you open an issue, make sure that you review these issues to see whether any of them present a quick solution for your problems:

Trouble summarizing one or multiple documents

Problem

The Content Assistant does not generate a summary for the selected documents.

Cause

You asked a question to the Content Assistant to generate a summary for the documents in your selection.

Solution

Use the watsonx ad hoc summary feature to generate a summary for one or multiple documents in your selection. For more information, see the topic Accessing the watsonx ad hoc summary for your document selection.

Trouble with questions that involve a mathematical calculation

Problem

The Content Assistant does not provide the results of a mathematical calculation from content across multiple documents.

Cause

The Content Assistant cannot draw data from across multiple pages, or perform many advanced calculations.

Solution

The Content Assistant can perform basic mathematical calculations if all the required data is in proximity. For example, a document can have the following sentence that says, "Jane had 5 apples and 4 oranges". The Content Assistant can provide the answer for a question that says, "How many pieces of fruit does Jane have?"

However, if the data is spread across multiple pages, the Content Assistant fails to find all the information and compute an answer. For example, consider that your selection contains insurance claim documents from car owners. The Content Assistant fails to provide an answer for a question that says, "How many hatchback cars met with an accident in the year 2025?"

User having trouble accessing the Content Assistant

Problem

The Content Assistant does not appear on the IBM® Content Navigator Desktop for a particular user.

Cause

You have reached the maximum number of active users limit for your subscription plan. For more information, see the topic Gathering Content Assistant usage information.

Resolution

Contact your IBM Sales representative to purchase an active user pack add-on for your subscription.

Trouble generating Content Assistant query responses

Problem

The Content Assistant does not generate answers for questions about your document content.

Cause

You have already reached the maximum number of questions limit for your subscription plan. For more information, see the topic Gathering Content Assistant usage information.

Resolution

Contact your IBM Sales representative to purchase a GenAI Questions pack add-on for your subscription.

Trouble generating watsonx summaries for document content

Problem

The Content Assistant does not generate watsonx summaries for your document content.

Cause

You have already reached the maximum number of watsonx summaries limit that can be generated for your subscription plan. For more information, see the topic Gathering Content Assistant usage information.

Resolution

Contact your IBM Sales representative to purchase a Document Summaries pack add-on for your subscription.

Trouble indexing document content

Problem

The Content Assistant fails to index the document content in your repository.

Cause

You have already reached the maximum indexed data size for your subscription plan. For more information, see the topic Gathering Content Assistant usage information.

Resolution

Contact your IBM Sales representative to purchase an Indexed Data Pack add-on for your subscription.

Trouble indexing and summarizing certain documents

Problem

The Content Assistant does not perform vector indexing and document summarization for certain documents. These documents remain unprocessed even though other documents are indexed and summarized successfully.

Cause

When you run a queue sweep for vector indexing and document summarization, the sweep assumes the identity of the object store administrator who installed the Content Assistant add-on for the object store. However, this object store administrator may not be part of user groups and roles that include READ and VIEW_CONTENT access to documents which the sweep processes. The problem occurs with Content Assistant deployments that use Content Platform Engine V5.7 and earlier only.

Resolution

For Content Assistant deployments using CPE V5.7 and earlier, you can set the following environmental variable:
-Dcom.filenet.engine.sweep.RelaxedQueueSecurity={76e57717-2d30-4198-8396-f88ec21e1daa},{8a920c77-ca2b-407f-9610-3a5ee8fd54a4}

Requesting your persisted data

From your document content, the IBM Content Assistant product stores extracted text excerpts and vector embeddings that are generated for the extracted text.

If you are terminating your subscription and you require an export of your persisted data, submit a request on the IBM Support page External link opens a new window or tab.

Didn’t find what you are looking for?

Check whether the problem is documented in the known issues section of the Content Assistant documentation.

If you need help with the Content Assistant, refer to the IBM Support page External link opens a new window or tab.