I'm getting very confused over looping and grouping in the data mapper (which I would be calling via a tHMap in a regular Data Integration job).

In essence we are receiving digital files for archiving, with a set of CSV metadata files which describe various aspects of the records. This metadata needs to be reformatted (joining data from across the original metadata files) and transformed, ultimately producing a set of JSON files, one per transferred digital file, so that we can send these (plus the digital files where appropriate - some files are restricted from public access for a set period so in that case we send only the metadata). Within this there can be multiple reasons why a file has public access restrictions, in the original metadata these are provided as a semicolon separated list within a single field. Independently of the access restrictions, there can also be a list of catalogue references to provide information about particularly close relationships between records. Similarly these are also provided as a semicolon separated list.

For the output JSON we have a JSON schema so I've created a structure from that (though this covers a wider range of use cases than is likely to occur in our particular usage, so some potential output fields wouldn't actually receive and data as we don't have appropriate data to put into them).

To try and create the input structures, I linked rows to the tHMap in the data integration job, the main row has normalised data with the list of reasons for an access restriction split out into individual values, with the other values repeated. There is an id value which acts as a key, so in the hierarchy these should be grouped together with the access restriction reasons going into an array.

Then there's a lookup with the normalised data for the relationships to other records, again these have the record id associated with each row in the lookup.

I think one problem may be that the structure for the output is based on a single record, but we have multiple records coming in on the other side, so I guess I either need to iterate over individual records somehow on the input side, or and a wrapper on the output, and then figure out how to split it up into individual JSON files, one per record later in the main job?

Then I can't figure out how to set up the looping and grouping so that this all comes together as required.

I can provide some screen shots of what I've got so far, or some (slightly simplified) example data which preserves the main points of what I've described above, but doesn't include every field in the actual data we're processing as there's rather a lot, and most of it doesn't have any bearing on the looping and grouping.

I hope this makes some sense: I'm a bit sleep deprived today too, which isn't helping.

Thanks

Hi,

Has anyone ever seen this error before? I encountered it on a logstream task:

00011524: 2026-10-04T03:00:01 [SORTER ]W: The record timestamp is at least 10 minutes later than the Replicate server time. Operation: BEGIN (6); Timestamp: Sat, 03 Oct 2026 17:00:47 GMT (1791046847000000); Replicate server time: Sat, 03 Oct 2026 16:00:01 GMT (1791043201849893); Stream position: 00000000.19489301.00000001.0000.00.0000:5773.491108.16 (sorter_transaction.c:204)

Thanks,

Mohammed

In Qlik Replicate version 2025.11.0.770 we are getting this error:

Hello everyone,

my department and I are currently bulding a new BI-Department.

At the moment I'm loading the TABLES from SF into Qlik. But I also have usecases where I have to concatenate several Tables.

One Engineer pointed out to my, why we're not using the views. So I tried to find usecased where people are using views and where the tables.

Unfortunately we have nobody in our team who is experienced in this kind of topic.

Is here anyone who can help me?

We are experiencing an issue with emoji rendering in the native Org chart visualization following our upgrade to Qlik Sense Enterprise on Windows November 2025, version 14.254.6.

Emojis that are correctly displayed in standard Qlik Sense tables are replaced by the Unicode replacement character � (U+FFFD) when displayed in the Org chart visualization, specifically in the card subtitle.

This issue affects all emojis we have tested.

• Product: Qlik Sense Enterprise on Windows

• Version: November 2025, 14.254.6

• Visualization: Org chart (Visualization Bundle)

• Browsers: Microsoft Edge, Firefox and Google Chrome

• The issue also occurs with newly created Org chart objects, so it does not appear to be limited to existing objects.

• It affects all emojis tested.

• The issue is reproducible across different computers and browsers.

• Resetting the Org chart styling does not resolve the issue.

• The Windows Segoe UI Emoji font is installed.

• Emojis are rendered correctly in standard tables within the same application.

• The issue appeared after upgrading Qlik Sense.

Any idea?

Hi All, 

I have created a session variable using below code

Our infrastructure security scanner flagged our Qlik Sense servers as potentially vulnerable because of the bundled Node.js 24.18.1 runtime. Following the October 21, 2026 Node.js security release, I tested updating that binary independently of a Qlik patch.

A scanner finding against a runtime version does not establish that the vulnerabilities are exploitable through Qlik Sense. Product-specific impact still needs confirmation from Qlik.

The update from Node.js 24.18.1 to 24.22.1 passed testing in our Qlik Sense Enterprise on Windows May 2026 Patch 7 environment. This is one environment's test result, not a compatibility guarantee.

This experimental method is not validated by Qlik. Qlik's documented approach is to update the bundled runtime through a Qlik Sense upgrade. This post shares a technical test, not an official remediation recommendation. If evaluating it, use development first, a maintenance window and a recoverable full backup.

Check the bundled version

& "$env:ProgramFiles\Qlik\Sense\ServiceDispatcher\Node\node.exe" --version

The full path checks Qlik's runtime rather than another Node.js installation on the PATH.

Run the attached scripts

1. Extract the scripts into one folder.
2. Download the official Node.js 24.22.1 Windows x64 binary.
3. Edit NodeUpdate.config.psd1: configure the binary, backup and log paths, and set the expected SHA256 using the win-x64/node.exe entry in SHASUMS256.txt.
4. Open Windows PowerShell 5.1 as administrator and run the backup:

.\01_Backup_Node.ps1

Continue only after BACKUP VERIFIED:

.\02_Update_Node.ps1

Leave the Service Dispatcher running before starting the update: the script handles stopping and restarting it. Execution is local to each machine; relative paths are resolved against the configuration folder.

Validate and roll back

Check Hub, QMC, application access, exports, reloads, usual connectors and Qlik logs. These are suggested checks, not a claim that all were covered in our test.

To roll back, set RestoreBackupDirectory to the exact original backup folder, then run:

.\03_Restore_Node.ps1

The scripts target the main Service Dispatcher runtime on Node.js 24 only. They do not update EAPowerTools, Package Cache or npm dependencies. A binary backup is not a full site backup, and automatic rollback is not guaranteed. Recheck the installed version after subsequent Qlik maintenance.

Sources: Node.js security bulletin, October 21, 2026; Node.js 24.22.1 release notes, October 21, 2026; Qlik Support guidance, updated August 14, 2024.

From https://www.aikido.dev/blog/node2shell-nodejs-brotli-rce-cve-2026-111497

Node.js shipped a security update on Wednesday for a flaw in how it decompresses Brotli data, a bug people are already calling Node2Shell (CVE-2026-111497). If an attacker can get a malicious payload in front of your app, the flaw can crash the process, leak part of its memory, or potentially let them run their own code.

Who is affected: anyone running Node.js or Bun whose app decompresses Brotli data that an attacker could control.

Ok, so Qlik Sense Enterprise on Windows runs on Node.js 24.14.1.
I have the following questions:

1. Does the specific implementation of a vulnerable Node.js version in Qlik Sense Enterprise on Windows implicate a threat to installations of Qlik Sense?

2. Are there any migitating actions that can be undertaken by administrators to continue operating safely until Qlik Sense/Node.js is patched?

3. Is there a patch coming? If so, when is this expected to be made available?

I have a table that currently works - it allows the user to select WeekStartShipDate, which is loaded in the load script with a Floor value, filtering the table to just one week:

The variables vStartSixWk and vEndSixWk determine the date range for the 6 Week column. 

vStartSixWk  =(WeekStartShipDate) -42

vEndSixWk  =(WeekStartShipDate)

The 6 Week column is then calculated as follows:

Sum({<WeekEndShipDate={">=$(vStartSixWk)<=$(vEndSixWk)"}>} EquivalentUnits) / 6

The % of 6Wk Total column is calculated as follows:

Num(
Sum({<WeekEndShipDate={">=$(vStartSixWk)<=$(vEndSixWk)"}>} EquivalentUnits) / 6
/
Sum(
TOTAL
Aggr(
Sum({<WeekEndShipDate={">=$(vStartSixWk)<=$(vEndSixWk)"}>} EquivalentUnits) / 6,
ShipTo
)
),
'##0.0%'
)

So, that all works fine until I add another dimension column, such as SalesOrder. Once I do that, the date filter breaks, and the 6 Week expressions return zeroes.

Please advise.