Product DocsMenu

Database Connector

The Coveo connector for databases allows to index database content directly through an ODBC connection and SQL queries.

The content of this section is also available in PDF format for download:
Database Connector Guide

Supported databases

The Database connector can index the following databases:

  • Microsoft SQL Server

  • Any database that has an OLE DB provider

  • Any database that has an ODBC provider, including Oracle and DB2

Connector Features

The features of the Database connector are:

Database indexing

Fully customizable database indexing, either by specifying SQL queries or item types (table names).

Incremental Refresh

The connector uses one column to track changes and only fetch the updated rows (see Enabling Incremental Refresh on a Database Source).

Note: A source full refresh or rebuild is required to take deleted items into account.

Authentication and identity replacement

The connector supports standard authentication to databases (username and password), however you can hide the password and username by introducing the @pwd and @uid tokens. By doing so, it is not possible to see passwords in clear text, even if one has access to CES Administration (see Replacing the Identity in Database Connection Strings).

Complement information retrieval using subqueries

The connector indexes documents through a query against a database. Subqueries can run on every document to complete the information with more complex queries (see Complement Information Retrieval with Subqueries for the Database Connector).

Support running in 32-bit

On a 64-bit platform, the 64-bit version of the connector can use ODBC providers in both 32-bit and 64-bit.

Query-based security provider

You can expand external groups and users using a security provider when the database contains information allowing groups to be expanded to external users and/or external users to be mapped to Active Directory users (see Enabling a Query-Based Security Provider for the Database Connector).

Paged query execution

The connector mapping file can be configured to support executing queries in smaller subsets of data, instead of one big result; therefore the time required to execute a query can considerably be reduced (see Enabling Paged Query Execution for the Database Connector).


The connector can run multiple threads, which can improve performances considerably (see Enabling Multithreading for the Database Connector).

Field mapping

Fully customizable field mapping, from your database fields to CES custom and system fields.


You have the possibility to pause and resume while indexing a database (see Enabling Pause/Resume on a Database Source).

Indexing custom permissions

The connector can index custom permissions predefined for each document indexed.

Redirecting hyperlinks

Capability to redirect the hyperlink of a document to an existing Web interface or its Quick View of a respective HTML cached version.

What's Next?

Get familiar with the steps to deploy the Database connector (see Database Connector Deployment Overview).