Prerequisites

The following section describes the client prerequisites when replicating to a Google Cloud BigQuery target.

Permissions

This section describes the Google Cloud BigQuery Service Account permissions required by Replicate.

When you create a service account key, a JSON file containing the connection information will be downloaded to your computer. You will need to copy the contents of this file to the Service account key field in the Google Cloud BigQuery endpoint.

For additional information about dataset creation, see Setting advanced connection properties.

Permissions required to create the target dataset automatically

If you want the dataset to be created automatically by Replicate, the following permissions are required:

BigQuery > BigQuery Job User

BigQuery > BigQuery Data Editor

Permissions required when target dataset already exist

If the dataset already exists, you need to perform the following steps:

1. Create a service account with the following permission:

   BigQuery > BigQuery Job User

2. Navigate to the dataset that you want Replicate to use, and then:

   1. Add the service account you just created as a principal.

   2. Assign the role: BigQuery Data Editor.

Drivers

Qlik Replicate on Windows

1. Download Simba ODBC driver version 3.0.0.1001 and install it on the Replicate Server machine:

2. Install Google Cloud SDK 402.0.0 on the Replicate Server machine as follows:
   1. Download and install the latest Google Cloud SDK version from Google.

   2. Downgrade to version 402.0.0 by issuing the following command:

      gcloud components update --version 402.0.0

Qlik Replicate on Linux

To be able to use Google Cloud BigQuery as a target endpoint in a Replicate task when Replicate is running on Linux, you need to perform the following procedure:

1. Download Simba ODBC driver version 3.0.0.1001 to the Replicate machine.

2. Extract:

   SimbaODBCDriverforGoogleBigQuery64_3.0.0.1001.tar.gz

   to:

   /opt/SimbaODBCDriverforGoogleBigQuery_3.0.0.1001-Linux

3. Move the files GoogleBigQueryODBC.did and simba.googlebigqueryodbc.ini to the lib directory under the Simba ODBC driver directory.

   Example:

   Move the files from:

   /opt/SimbaODBCDriverforGoogleBigQuery_3.0.0.1001-Linux/setup

   to:

   /opt/SimbaODBCDriverforGoogleBigQuery_3.0.0.1001-Linux/SimbaODBCDriverforGoogleBigQuery64_3.0.0.1001/lib

4. Edit the simba.googlebigqueryodbc.ini file as follows:

   1. Change the ErrorMessagesPath to the path of the XML file containing the ODBC messages. The default location is:

      ErrorMessagesPath=/opt/SimbaODBCDriverforGoogleBigQuery_3.0.0.1001-Linux/SimbaODBCDriverforGoogleBigQuery64_3.0.0.1001/ErrorMessages

   1. Change the DriverManagerEncoding to UTF-16.

5. Add the following path to the site_arep_login.sh file located in the Replicate bin directory:

   export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/SimbaODBCDriverforGoogleBigQuery_3.0.0.1001-Linux/SimbaODBCDriverforGoogleBigQuery64_3.0.0.1001/lib

6. Edit the /etc/odbcinst.ini file and add the driver path (i.e. the path where the driver is installed):

   [ODBC Drivers]

   Simba= Installed

   Simba ODBC Driver for Google BigQuery = Installed

   [Simba ODBC Driver for Google BigQuery]

   Description=Simba ODBC Driver for Google BigQuery(64-bit)

   Driver=/opt/SimbaODBCDriverforGoogleBigQuery_3.0.0.1001-Linux/SimbaODBCDriverforGoogleBigQuery64_3.0.0.1001/lib/libgooglebigqueryodbc_sb64.so

7. Change to the Replicate bin directory (<install_dir>/bin). Then stop and start the Replicate service:

   ./areplicate stop
   ./areplicate start

8. Download and install Google Cloud SDK 402.0.0 for Linux on the Replicate machine.