The Qlik ODBC Connector Package log analyzer app is an easy-to-set-up app meant to assist in troubleshooting ODBC Connector issues for Qlik Sense Enterprise on Windows (on-prem).

This short guide documents how to configure and use the app in your own environment for both analysis and learning purposes. 

Content

• Requirements
• Setup
• App Content

Requirements

To get as much information as possible and leverage the app at its fullest, set the following two log settings in the QvOdbcConnectorPackage.exe.config file. The file is typically found in C:\Program Files\Common Files\Qlik\Custom Data\QvOdbcConnectorPackage.

• LoggingLevel set to ALL
• LogSQLStatements set to true
  
  

Setup

1. Download the Qlik ODBC Connector Package log analyzer app, either from the article directly or from our file share. 
2. Import the Qlik ODBC Connector Package log analyzer app to your Qlik Sense environment
3. Open the app from the hub
4. Go to the Data Load Editor 
5. Open the config tab
6. Define a lib connection for your ODBC connector package log file
   
   
   
   
   
7. Reload the Qlik ODBC Connector Package log analyzer app

App Content

The first sheet opens with aggregated information and KPIs.

What you will find here:

• Connector Package version
• ApplicationId: The applications using the ODBC Connector Package connectors
• Logging Level: The applied ODBC Connector Package logging level
• HostName
• Start time: The start timestamp of the logging
• End time: The end timestamp of the logging
• Date: The logged dates
• UserId: The logged user IDs
• Count of errors and Count of warnings: The number of logged errors and warnings
• Connector activity during the day: A chart showing during what time ODBC connectors have been used
• Connector activity by days: A chart showing during which days ODBC connectors have been used
• Max query data size reloaded: KPI showing the maximum data size reloaded by an executed query within a logged reloadID
• Max query reload time: KPI showing the maximum time of all the logged and executed queries
• Data source usage: A table showing the count of connections, count of used apps, count of errors, and the sum of the data size of all reloaded data by data source
• Command pipe: A list of all command pipes. Command pipes are the basis for an ODBC Connector Package activity. A command pipe is created every time the Data Load Editor is opened and the ODBC connections are loaded, as well as when a new ODBC connection is created, if a connection test is performed, and when data is previewed or reloaded. 
• RequestId: A command pipe can have different kinds of request IDs. For more details, see the Connector process details section.
• Num command pipes: The number of command pipes
• Number of apps: The total number of apps using some form of ODBC connection

The first sheet also shows additional details supplementing the initial charts and KPIs.

What you will find here:

• Connector process details: Filtering by a specific Command pipe allows for an overview of all messages and requests for that specific command pipe sorted by timestamp
• Database and connection string details: A table showing the type of data source, database version (if available), and the used connection string
• Overview of reload executions: A table showing which SQL statements have been executed in which request and command pipe. Also contains other details, such as timestamps, query time, reloaded data size, data source, rows fetched, etc...

The second sheet begins with a summary of errors and warnings.

What you will find here:

• A list of errors, warnings, and exceptions by command pipe and sorted by timestamp.
• The exact message and which app has been impacted

Following the error analysis, the app lists information about Advanced parameters and connection details.

What you will find here:

• The connection details are broken down by connection string parameters and can be filtered by command pipe. This allows you to review which reload used which connection string parameters and advanced parameters.

The last section of the second sheet displays table field details.

What you will find here:

• Data type information by executed SQL query, command pipe, application ID, and data source.

The Qlik ODBC Connector Package log analyzer app is provided as-is. No ongoing support will be provided by Qlik Support. For questions and to engage with your active Qlik peers on the subject, head over to Management & Governance.

Environment

• Qlik ODBC Connector Package
• Qlik Sense Enterprise on Windows

This article explains how to change the log level for the QvWebStorageProviderConnector in Qlik Sense Enterprise on Windows.

Always back up .config files before making changes.

To change the log level of your QvWebStorageProviderConnectorPackage to debug mode:

1. Browse to %Program Files%Common Files\Qlik\Custom Data\QvWebStorageProviderConnectorPackage
2. Locate deploy.config
3. Open deploy.config file in a text editor with elevated permissions (Administrator)
4. Locate the <LoggingLevel> section:
   

   <!-- Valid values: Off, Error, Warn, Info, Trace - We would recommend you have at least Error set.
   --> 
   <LoggingLevel>Info</LoggingLevel>

5. Change the field Info to Trace
   
   Example:

   <LoggingLevel>Trace</LoggingLevel>

6. Save the file
7. Restart all Qlik Sense services

Environment

• Qlik Sense Enterprise on Windows

This capability is being rolled out across regions over time:

1. May 5: India, Japan, Middle East, Sweden
2. TBD: Asia Pacific, Germany, United Kingdom, Singapore
3. TBD: United States
4. TBD: Europe
5. TBD: Qlik Cloud Government

[Update: May 7] The previously-scheduled rollouts of Automation Sharing and Collaboration for some regions have been temporarily postponed. We are working on an updated release plan and updated release dates are soon to be determined (TBD). Thank you for your understanding.

With the introduction of shared automations, it will be possible to create, run, and manage automations in shared spaces. 

Content

• Allow other users to run an automation
• Collaborate on existing automations
• Collaborate through duplication
• Extended context menus
• Context menu for owners:
• Context menu for non-owners:
• Monitoring
• Administration Center
• Activity Center
• Run history details
• Metrics

Allow other users to run an automation

Limit the execution of an automation to specific users.

Every automation has an owner. When an automation runs, it will always run using the automation connections configured by the owner. Any Qlik connectors that are used will use the owner's Qlik account. This guarantees that the execution happens as the owner intended it to happen.

The user who created the run, along with the automation's owner at run time, are both logged in the automation run history.

These are five options on how to run an automation:

1. Run an automation from the Hub and Catalog
   
   
   
   
2. Run an automation from the Automations activity center
   
   

    

3. Run an automation through a button in an app
   
   You can now allow other users to run an automation through the Button object in an app without needing the automation to be configured in Triggered run mode. This allows you to limit the users who can execute the automation to members of the automation's space.
   
   More information about using the Button object in an app to trigger automation can be found in How to run an automation with custom parameters through the Qlik Sense button.
   
   
4. Programmatic executions of an automation
   
   
   1. Automations API: Members of a shared space will be able to run the automations over the /runs endpoint if they have sufficient permissions.
   2. Run Automation and Call Automation blocks
      
      
5. Note for triggered automations: the user who creates the run is not logged as no user specific information is used to start the run. The authentication to run a triggered automation depends on the Execution Token only.

Collaborate on existing automations

Collaborate on an automation through duplication.

Automations are used to orchestrate various tasks; from Qlik use cases like reload task chaining, app versioning, or tenant management, to action-oriented use cases like updating opportunities in your CRM, managing supply chain operations, or managing warehouse inventories.

Collaborate through duplication

To prevent users from editing these live automations, we're putting forward a collaborate through duplication approach. This makes it impossible for non-owners to change an automation that can negatively impact operations.

When a user duplicates an existing automation, they will become the owner of the duplicate. This means the new owner's Qlik account will be used for any Qlik connectors, so they must have sufficient permissions to access the resources used by the automation. They will also need permissions to use the automation connections required in any third-party blocks. 

Automations can be duplicated through the context menu:

As it is not possible to display a preview of the automation blocks before duplication, please use the automation's description to provide a clear summary of the purpose of the automation:

Extended context menus

• Start a run from the context menu in the hub
• Duplicate automation
• Move automation to shared space
• Edit details (owners only)
• Open in new tab (owners only)

Context menu for owners:

 

Context menu for non-owners:

 

 

Monitoring

The Automations Activity Centers have been expanded with information about the space in which an automation lives. The Run page now also tracks which user created a run.

Note: Triggered automation runs will be displayed as if the owner created them.

Administration Center

The Automations view in Administration Center now includes the Space field and filter.

The Runs view in Administration Center now includes the Executed by and Space at runtime fields and filters.

Activity Center

The Automations view in Automations Activity Center now includes Space field and filter.

Note: Users can configure which columns are displayed here.

The Runs view in the Automations Activity Center now includes the Space at runtime, Executed by, and Owner fields and filters.

In this view, you can see all runs from automations you own as well as runs executed by other users. You can also see runs of other users's automations where you are the executor.

Run history details

To see the full details of an automation run, go to Run History through the automation's context menu. This is also accessible to non-owners with sufficient permissions in the space.

The run history view will show the automation's runs across users, and the user who created the run is indicated by the Executed by field.

Metrics

The metrics tab in the automations activity center has been deprecated in favor of the automations usage app which gives a more detailed view of automation consumption.

This article addresses a common concern where users observe frequent ROLLBACK operations in the logs when using IBM DB2 LUW as a source in data replication or change data capture (CDC) environments. These messages may appear alarming at first, but they are typically part of the database's normal transaction handling process.

Understanding why these rollbacks occur and how they impact system performance can help reduce confusion and improve monitoring strategies.

Content

• Prerequisites
• Step-by-Step Instructions
• Step 1: Understand the Nature of ROLLBACK Operations
• Step 2: Why You See Frequent ROLLBACK Messages
• Step 3: Impact on Performance and Latency
• Step 4: Best Practices for Monitoring and Optimization
• Common Issues and Solutions
• Conclusion
• Related Content

Prerequisites

Before proceeding, ensure the following:

• Basic understanding of database transactions and replication concepts.
• Access to:
  • DB2 LUW logs
  • Replication
  • Administrative privileges on the source database (for log access and configuration)

Step-by-Step Instructions

Step 1: Understand the Nature of ROLLBACK Operations

• A ROLLBACK is a standard database operation used to undo uncommitted changes.
• In systems like Order Entry, a transaction may be initiated (e.g., adding an order) but not committed.
• If the transaction is later canceled, DB2 issues a rollback to revert the changes.

Step 2: Why You See Frequent ROLLBACK Messages

• These messages are generated when Qlik Replicate captures rollback events from the DB2 transaction logs.
• Rollbacks are especially common in high-transaction environments where temporary or test entries are frequently created and discarded.

Step 3: Impact on Performance and Latency

• Rollbacks can increase the volume of log data that needs to be processed.
• This may introduce latency in replication tasks, especially if:
  • Many transactions are rolled back in a short time.
  • The rollback spans multiple tables or large datasets.
• Additional transaction logs generated during rollback can also slow down log reading and processing.

Step 4: Best Practices for Monitoring and Optimization

• Monitor the frequency of rollbacks using DB2 monitoring tools or SQL queries on the transaction log.
• Review application logic to ensure transactions are committed or rolled back efficiently.
• If rollbacks are excessive and unexpected, investigate application behavior or consider tuning transaction handling.

Common Issues and Solutions

Issue: High latency in replication tasks
Solution: Check for excessive rollback activity and optimize transaction handling in the source application.

Issue: Concern over data integrity due to rollbacks
Solution: Rollbacks are part of the normal process and ensure consistency. No action is needed unless they are unusually frequent.

Issue: Large volume of logs generated
Solution: Consider adjusting log retention policies or archiving strategies to manage disk usage.

Conclusion

Frequent ROLLBACK messages in DB2 LUW are typically a normal part of transaction management, especially in dynamic systems with frequent data changes. While they can contribute to replication latency, they are essential for maintaining data integrity. Monitoring and understanding these operations can help you better manage system performance.

Related Content

• IBM DB2 LUW Documentation
• IBM CDC (Change Data Capture) Overview
• Best Practices for DB2 Transaction Logging

Environment

• Qlik Replicate

A Qlik Replicate task fails with error

Query timeout expired

Resolution

The error occurs when Qlik Replicate is working on a query to the source or target and the query takes longer than anticipated, leading to the timeout. 

Edit the source or target endpoint and within the Advanced properties and add a new Internal Parameter named cdcTimeout.

1. Stop the task.
2. Open source or target endpoint depending on the error from the query source or target and add the following Internal Parameter:
   
   cdcTimeout (default 600); (some endpoints do not have this value)
   executeTimeout (default 60)
   loadTimeout (default 1200).
   
   Try 10x the default value. Values are in seconds.
   
   See Qlik Replicate: How to set Internal Parameters and what are they for? for more information.
   
   
3. Save the endpoint
4. Try to resume task and see if we still have a timeout issue.

Environment

Qlik Replicate 

You might be encountering the following issue on the Talend Administration Center UI (Job Conductor page) when deploying a task after upgrading or applying a Talend Administration Center patch

Error: ERROR ExecutionTaskHandler - /Talend/Administrator/generatedJobs/task****.zip (No such file or directory)

Resolution

Please verify the 'Generated Jobs' folder on the Talend Administration Center Application

• If the folder is missing, create the appropriate path and configure it in the Job Conductor configuration page.
• If the folder exists but lacks the necessary permissions, ensure that write permissions are granted.

Cause

This error occurs when the path to the 'Generated Jobs' folder configured in the Job Conductor is missing or has insufficient permissions.

Environment

Talend Administration Center 

This article briefly introduces how to request SOC 2 audit report from Stitch

As Stitch has been certified compliant with the SOC 2 (System and Organization Controls 2) security, availability, and confidentiality principles by an independent auditor, the audit report can be requested by contacting the Sales team.

You will need to fill out the form below, and our Sales Team will connect with you to obtain the SOC 2 report.

https://www.stitchdata.com/contact/

If you are not getting a response on time, please feel free to contact Qlik Support to get your account representative's details How to contact Qlik Support

Related Content 

For more information about security issues, please refer to below link
https://www.stitchdata.com/docs/security/overview#security-issues

Environment

Talend Cloud 

The PostgreSQL instance used by Qlik Sense (on-premise) uses 100% of CPU capacity. This behavior can be seen in an interval of 2 hours.

Cause

The Resource Distribution Service is set to run a 2-hourly scan on all extensions and themes as part of a sync to Qlik Sense Cloud. This has proven to have a dramatic effect in very large environments where postgres.exe is causing the CPU to spike to 100% for an extended duration, making the environment unusable.

Resolution

Switch off Resource Distribution or set it to run every 24 hours.

Switch off Resource Distribution

1. Open the services.conf file (default: C:\Program Files\Qlik\Sense\ServiceDispatcher\services.conf) in a text editor elevated to Administrator permissions
2. Locate the [resource-distribution] section.
3. Set Disabled to true:
   

   [resource-distribution]
   Disabled=true
   Identity=Qlik.resource-distribution
   DisplayName=Resource Distribution
   ExePath=Node\node.exe
   Script=..\ResourceDistributionService\server.js

4. Save the file
5. Restart the Service Dispatcher

Schedule it to run every 24 hours

1. Open the services.conf file (default: C:\Program Files\Qlik\Sense\ServiceDispatcher\services.conf) in a text editor elevated to Administrator permissions
2. Locate the [resource-distribution] section.
3. Set Disabled to false as //Disabled=true:
   

   [resource-distribution]
   //Disabled=true
   Identity=Qlik.resource-distribution
   DisplayName=Resource Distribution
   ExePath=Node\node.exe
   Script=..\ResourceDistributionService\server.js
   
   [resource-distribution.parameters]
   --secure
   --wes-port=${WESPort}
   --mode=server
   --log-path=${LogPath}
   --log-level=info
   --interval-time=86400000
   

    

4. Save the file
5. Restart the Service Dispatcher

Internal Investigation ID(s)

QB-18723

Environment

Qlik Sense Enterprise on Windows (November 2022 and later)

A binary load command that refers to the app ID (example Binary[idapp];) does not work and fails with:

General Script Error

or

Binary load fails with error Cannot open file

This is observed after upgrading to Qlik Sense Enterprise on Windows November 2024 Patch 8, and any later releases.

It was previously possible to do binary loads from app binaries without the .qvf file extension (GUID number file name). With recent releases, Qlik is enforcing an already existing restriction, meaning the QVF file extension is now required. See  Binary Load and Limitations for details.

This is working as designed and is intended as a security measure.

Example of a valid binary load:

LIB CONNECT TO 'DataFiles';
  LOAD * FROM [YourFile.qvd];

Example of an invalid binary load:

"Binary [lib://Apps/777a0a66-555x-8888-xx7e-64442fa4xxx44];"

Environment

• Qlik Sense Enterprise on Windows November 2024 Patch 8 and any later releases

In many cases, when using a dynamic schema, it's necessary to change the properties of a column to map the target data structure. The column properties include column name, value, type, etc.

Example Job

The example Java code below changes the column name of column 'name' to a new name, changes the data type of column 'age' to Integer. Copy this code and modify it according to specific needs, it supports changing column name, data type, length, precision, etc.

Dynamic columns = input_row.data;
for (int i = 0; i < columns.getColumnCount(); i++)
{
DynamicMetadata columnMetadata = columns.getColumnMetadata(i);
String col_name=columnMetadata.getName();
String col_type=columnMetadata.getType();

if(col_name.equals("name")){
columnMetadata.setName("new_name");
}

if(col_name.equals("age")){
columnMetadata.setType("id_Integer");
}
}
output_row.data=columns;​

You may be experiencing the following data loading error when connecting the HubSpot Integration to a destination:

"The data for this table would result in more than 1,600 columns" 

Resolution

To resolve this issue and bring your integration back online, follow the steps below. This process applies to any affected table (for example, contacts, deals, companies tables)

1. Pause the Integration
   Start by pausing the HubSpot integration in Stitch to prevent new data from syncing during the fix.
2. Update Field Selection
   • Navigate to the affected table (for example: contacts, deals) in the Stitch UI.
   • Deselect the main properties field.
   • Manually select only the individual fields you need — these will be listed as properties_<field_name>
3. Drop the Affected Table in the destination( Redshift as an example)

   In your database, drop the oversized table to allow Stitch to recreate it with the updated schema.

4. Reset the Table in Stitch
   • Go to Tables to Replicate > [table_name].
   • Click on Table Settings.
   • Choose Reset Table, and confirm the prompt.

   This will trigger a rebuild of the table with the selected fields only.

5. Unpause the Integration

   Once everything is updated, unpause the integration to resume syncing.

Cause

This error message indicates that the data being replicated into your destination exceeds the 1600-column limit during loading imposed by the warehouse (e.g, Redshift) and it most commonly arises when Stitch attempts to de-nest complex fields like properties from HubSpot tables, such as: contacts, deals,companies

These properties fields act as containers for a large number of individual fields. When de-nested, they can generate a lot of columns, especially if:

• The top-level properties field is selected, and

• Individual properties_<field_name> fields are also selected

Even if you deselect some individual fields, selecting the main properties field overrides those exclusions, and all properties are still replicated. This can push your table schema past the 1600-column limit.

Environment

Talend Cloud 

Question I

What is Community Supported Integrations in Stitch?

Community Integrations are developed by a member of the Singer open source community and the code to run it is publicly available.

While Singer Community Integrations can be run within Stitch, we does not actively support them. The Singer Community will maintain and Support Community integrations. This includes fixing bugs and adapting to new versions of third-party APIs.

Question II

How to Request Support for Community Supported Integration?

To request support for this community-supported integrations regarding any issues, you have a few options forward, and they are outlined below:

1. You can open an “issue” on the integration repo
2. Reach out to the Singer Community via the Singer Slack group. Provide relevant Log lines from the Extraction Logs page within Stitch and a brief summary of the behavior, as well as any additional context they have surrounding its occurrence will be helpful for any community members who move to address the issue, and our Engineering team also periodically review issues and pull requests opened on Singer repositories.
3. Contribute to the code yourself by creating a pull request. With all of that mentioned, since the error is due to a forbidden connection, you should make sure that nothing related to permissions has changed, and try to re-authorise the integration and then run a new job. You should also try generating a new API Token here:
   https://www.stitchdata.com/docs/integrations/saas/shiphero#setup Singer | Open Source ETLSimple, Composable, Open Source ETL 

This information is only related to the community-supported integrations.

Environment

Talend Cloud 

Starting from Qlik Replicate versions 2024.5 and 2024.11, Microsoft SQL Server 2012 and 2014 are no longer supported. Supported SQL Server versions include 2016, 2017, 2019, and 2022. For up-to-date information, see Support Source Endpoints for your respective version.

Attempting to connect to unsupported versions, both on-premise and cloud, can result in various errors.

Examples of reported Errors:

• Qlik Replicate 2024.5 accessing Microsoft Azure SQL Server 2014
  
  [SOURCE_CAPTURE  ]W:  Table 'dbo'.'tableName' has encrypted column(s), but the 'Capture data from Always Encrypted database' option is disabled. The table will be suspended  (sqlserver_endpoint_capture.c:157)
  
  
• Qlik Replicate 2024.11 accessing Microsoft SQL Server 2014
  
  [SOURCE_UNLOAD ]E: RetCode: SQL_ERROR SqlState: 42S02 NativeError: 208 Message: [Microsoft][ODBC Driver 18 for SQL Server][SQL Server]Invalid object name 'sys.column_encryption_keys'. Line: 1 Column: -1 [1022502] (ar_odbc_stmt.c:4067)

Cause

The system view sys.column_encryption_keys is only available starting from SQL Server 2016. Attempting to query this view on earlier versions results in errors.

Reference: sys.column_encryption_keys (Microsoft Docs)

Resolution

Upgrade your SQL Server instances to a supported version (2016 or later) to ensure compatibility with Qlik Replicate 2024.5 and above.

Internal Investigation ID(s)

00375940, 00376089

Environment

• Qlik Replicate versions 2024.5, 2024.11, and higher
• Microsoft SQL Server 2014, 2012, and lower (unsupported)

Setting up a Qlik Cloud Analytics connection to an Azure SQL database with Microsoft Entra MFA enabled fails at the connection test with:

Error message: Please check the value for Username, Password, Host and other properties.
Description: Communication link failure

ERROR [08S01][Qlik][SqlServer] Connection broken unexpectedly.

Resolution

Create an Azure SQL database connection using a supported authentication method.

See Authenticating the driver.

Microsoft Entra MFA is not supported.

Environment

• Qlik Cloud Analytics

After otherwise successfully restoring a Qlik NPrinting backup, the license information remains blank. This information is expected to be restored as well.

Resolution

This has been addressed as defect QCB-31631.

Workaround:

Activate the Qlik NPrinting license manually.

Fix Version:

Qlik NPrinting February 2024 SR5 and later releases. See the Release Notes for details.

Environment

• Qlik NPrinting May 2023 SR6 and SR7
• Qlik NPrinting February 2024 SR3 and SR4

Information provided on this defect is given as is at the time of documenting. For up-to-date information, please review the most recent Release Notes or contact support with the ID QCB-31631 for reference.

Due to tHttpRequest, tFileFetch and tRest components have been removed from Talend Studio 8.0 R2025-04 onwards in favor of tHTTPClient, some adjustments are required to transition to tHTTPClient if your current Talend jobs actively use the deprecated  tHttpRequest/tFileFetch/tREST/component.

This article briefly introduces how to find all jobs rely on the deprecated components for replacement purpose.

How to

1. Create a dummy job or find a Job using the related deprecated components
2. Right click on the component and choose " Find Component in Jobs" item from List and it will search all jobs that use this component.SearchComponentUsedinJobs 

Internal Investigation ID(s)

QTDI-783 - Deprecate tREST component in favor of tHTTPClient
QTDI-784 - Deprecate tFileFetch component in favor of tHTTPClient
QTDI-785 - Deprecate tHTTPRequest component in favor of tHTTPClient

Environment

Talend Studio 

Visualizations calling data points within a Qlik App from Qlik Predict (previously AutoML) produce the following error:

Error: Operation was rejected because the system is not in a state required for the operation's execution.

An additional error seen may be:

"grpc::StatusCode::FAILED_PRECONDITION: 'Error returned by endpoint: InternalServerError Internal Server Error'".

Resolution

Retrain the model with the same training data last used to train the Qlik Predict, then reload the app.

Cause

Qlik has recently performed back-end changes that may impact a small number of tenants which have not had their Qlik Predict models retrained since March or April 2025. 

Internal Investigation ID(s)

SUPPORT-1294

Environment

• Qlik Predict

The central proxy virtual proxy cannot be deleted and virtual proxy prefix are unique per proxy.

Environments:

• Qlik Sense Enterprise on Windows 

Resolution:

In order to change to remove the prefix for a different virtual proxy to use another authentication method by default:

1. in the QMC, go to "Virtual proxies" > "Central proxy (default)"
2. Modify the virtual proxy prefix to the desired string
3. Click "Apply" in order to save
4. Go back to "Virtual proxies", double-click on the virtual proxy that you want to make default in order to edit it.
5. Remove the virtual proxy prefix
6. Click "Apply" in order to save

Note: If the virtual proxy made default is using SAML authentication, the SAML Assertion Consumer Service URL will also need to be updated in the Identity Provider configuration.

Some data connections (such as the Microsoft SQL connector in the ODBC Connector package) may stop working after the Qlik Sense on-premise service account password is changed.

The following error is shown:

LIB CONNECT TO [Connection-Name]
Error: Connection aborted (System error: The handle is invalid.)
Execution Failed
Execution finished.

This happens even if the user ID for the connection itself is not the same as the one used for the service account.  

Resolution

Re-created encryption keys. For details, see Solution 3 in Qlik Sense Data Connectors are missing.

Environment

• Qlik Sense Enterprise on Windows