---
title: "Connect and monitor knowledge sources"
description: "Distinguish stored files, successful sync and searchable content."
url: https://helpcenter.complytrain.net/en/articles/195-connect-and-monitor-knowledge-sources
collection: "Organisation knowledge"
updated: 2026-09-22
language: en
---

# Connect and monitor knowledge sources

Distinguish stored files, successful sync and searchable content.

Distinguish stored files, successful sync and searchable content.

## Add a source

Product location: `/tenant-context/sources`

Open **Knowledge Sources**. Use quick upload for files, a web crawler for public pages, an S3 source for eligible objects or the SharePoint connection flow where available.

Describe uploaded material and select the actual document type; it determines workflow and retention.

For a web crawler, enter a **Source Name** and, optionally, a **Description**. Under **Connection Configuration**, list the starting addresses in **Seed URLs (one per line)**: at most 50, each starting with http:// or https://. Choose the **Crawl Scope**: **Same host only**, **Same domain (includes subdomains)** or **Any linked pages**. **Max Pages** is optional; when you set it, it must be a whole number from 1 to 10000.

An Amazon S3 source's connection holds the **S3 Bucket Name**, a **Key Prefix (optional)** that limits syncing to object keys starting with it, the **AWS Region** and an **IAM Role ARN (optional)** for a cross-account role. The bucket name and region cannot be blank or consist only of spaces.

Under **Sync Settings**, **Enable automatic sync** syncs the source at the chosen **Sync Frequency**: **Real-time (webhook)**, **Hourly**, **Daily**, **Weekly** or **Manual only**.

A value that cannot be accepted is named under its field, for example **Enter a whole number of pages between 1 and 10000**, **A crawl source can start from at most 50 addresses** or **Every line must be a web address starting with http:// or https://**. Correct it before saving; the same limits are checked again when the source is saved. To remove a description, clear the box and save; the source is stored with no description.

## Choose indexing and destination

Product location: `/tenant-context/sources`

Review **Auto-index to knowledge base** and the destination vault. The selected vault also affects which knowledge base makes the documents searchable. Stored documents can require Document Vault access to open even when the source list is visible.

A sync can download a file successfully before AI indexing finishes. Inspect **Ingested files** and each file's indexing reason when content does not appear in answers.

## Maintain the source

Product location: `/tenant-context/sources`

Open **Sync history** to inspect queued, running, completed or cancelled jobs and any reported failure. Check the scope and processing state before requesting another sync.

Pausing a source stops future syncing while retaining already-synced documents. Removing or changing a source connection should not be assumed to remove every previously stored document; inspect the actual vault and source confirmation.
