TimeXtender Desktop Q&A

Hello,TimeXtender 20.10.31.64ODX: 20.10.31We are currently running a database on hyperscale with 12vcores and premium memory optimized hardware. ODX is a data lake.We are running execution packages on 4 threads and we want to reduce our odx transfer times as that is out biggest impact right now.This is an example of a load. I am seeing long ODX transfer times for very small tables that I cant find the cause of or explain.We have alot of small tables (see example below) that is taking 1-3 minutes, while some tables doing data cleansing on 1million + rows is doing it in the same amount of time. See this small table that is 400 rows, no transformations and only 10 columns. I can see we are peaking log io at 100% during certain periods.However, that seems to be during the later parts of the loads and not during odx transfer in the start. Is there any thing that can cause this or possibility to improve it?Thank you,Victor​​​​​

Hi Everyone, In the Ingest instance activity log I see a lot of messages. It happens when populating a table in the DSA (execute). We are using Datalake as storage and TDI version 7158.1Execution takes a loooooong time also. What goes wrong here???  

Hi team, I want to create a new enviroment variable that I used to do in the old version, but with new version it changed. The idea that create a new enviroment varible, and configure on execute package, the variable check dev enviroment or prod enviroment and if it prod the execute package will run. My Version isHow can I do that?

Hello,I use TimeXtender CSV Data Source (version 23.18.5.0). I'm trying to read about 15 csv's from an sftp server, but since january 15th 2026 we get an error. This server is hosted by an external supplier which has it's settings limited to max 1 session (no concurrent sessions).In TDI I've configured the max concurrent threads on 1, so max 1 thread at a time. My transfer still crashes because I use too many threads.I tried to put max concurrent threads on 0. This resulted in a never ending transfer task. Didn't work.After this the supplier increased it's max threads temporarily for testing purposes. I configured the connector on 1 thread again. If I look in the log, I can see that the first 2 tables start on the same timestamp. This occurs a couple of times more. Is the connector using 2 threads now?Is this advanced max thread setting working correctly for the TX csv connector? Supplier claims we're not using 1 thread. I'm not sure how I can check and prove this. Do you have any advi

We have recently updated a TimeXtender version from 20.1047 to 20.10.74 and was successful for Development and QA but after updating Production the SSIS Packages are not executing. Development and QA have the same version of SQL as Production. The execution of SSIS Packages is setup through External tasks. The only relevant information I have found so far regarding this issue is here (https://legacysupport.timextender.com/hc/en-us/articles/360000169786-Troubleshooting-Upgrade-Issues) but am not sure how the steps under heading ‘Cannot deploy or execute SSIS packages’ would resolve the issue. Anyone else run into this?

Hello,Since the TimeXtender Classic (V25) was announced to be released by the end of Q2 2025, i would like to ask if there is a clarified date when the release is planned. To my knowledge it has not been released as of yet? Also the more important question to me is the upgrade process. Currently we are using TimeXtender classic together with ODX server and we have couple of questions:Will there be a detailed guide released on upgrading older TimeXtender platforms (20.10.52 in our case) to Classic V25? What will happen with all the data sources in our ODX instance? Since we are using quite some CDATA providers - does it mean we will have to manually recreate the data sources? Or ODX will be deprecated and we need to move to using BusinessUnits? Thanks for the information

I have a source table in ODX. The source only keeps record for the last 3 months. UUID ColumnA ColumnB ColumnC Date 1234 Xiaomi Mobile phone 125 GB 01.01.2025 5678 Apple iPad 1 TB 01.02.2025 9101 Samsung Mobile phone 256 GB 01.03.2025  When i am in month 4, the forst row is unavailable in the sourceUUID ColumnA ColumnB ColumnC Date 5678 Apple iPad 1 TB 01.02.2025 9101 Samsung Mobile phone 256 GB 01.03.2025 11121 Google Pixel Mobile phone 256 GB 01.04.2025  I want to keep the history in DSA forever.I know that if i create a history table and use UUID as unique key, i will be able to keep the history.The question is: what if i accedently execute the table on 01.04.2025! Will i lose everything before 01.02.2025? Or will executing the table keep the history intact !What if there are changes in the source and there is either a new column or the column name changes? I need to then deply the table. Will that keep th

Hello, we have upgraded to 20.10.74 TimeXtender with 20.10.73 ODX server from older version. Since ODX server is getting removed in the future and CData connectors have their distribution stopped, i am looking to migrate the existing connectors to future supported ones and  switching from using ODX server to using Business Unit. Could you advice me which connectors i should start using instead of the following ones?:CData ADO.NET Provider for Microsoft Excel Online 2024; CData ADO.NET Provider for Databricks 2024; CData ADO.NET Provider for SAP HANA 2023; CData ADO.NET Provider for REST 2023I does not look that there are direct alternatives to these, at least in current TX version. Are there also any guides of migrating these specific ones to new alternatives? Thank you

We are loading data from a SQL Server (onprem) to a Datalake (ingest instance). Data is stored as parquet, but TX is creating a huge parquet file for each table.  We are using ADF to ingest the data into the ingest instance.  Is there a way to load the data in such a way that when it gets into the Datalake the parquet can be divided in parts?  table_xxx_0001.parquet, table_xxx_0002.parquet, etc. (I know this is not the parquet format it is just an example).  Is this possible? We are having performance issues reading from parquet to AZURE SQL DB prepare instance.  We are using ADO.NET from ingest to prepare.

Hi, I am trying to conect to an endpoint using a dynamic query parameter from a simple SQL query: SELECT       [CODE_ID] as codeId  FROM [dsa].[Products]I use {codeId} as value in the query parameter. When deploying, I get the following error: Format of the initialization string does not conform to specification starting at index 0.  When I hard-code the codeId, it works. Using this query does not work. There are around 600 codeId's in the query. Even when i limit it to 2 id's, it gives me the same error. My API documentation does not mention anything about extra parameters. Also, the logs do not give me much more information, other than the error I mentioned. Anyone has experience with this error and has a possible solution? Kind regards,René 

Hi,We are trying to access ODX execution history (backlog / queue / duration / errors) for log analytics purposes.We understand from documentation and release notes that the local ODX backlog (Backlog.sqlite) has been removed / moved in newer ODX versions, and that execution metadata is now handled differently.However, we are currently unable to locate where this information can be accessed programmatically.Additionally, in the TimeXtender Portal, we do not see the menu optionAdmin → Security / Permissions, so we are unable to: verify roles/permissions access or inspect the API for execution data Could you please help clarify: Where ODX execution history (status, start/end time, duration, errors) is stored in newer ODX versions Whether this data is accessible via: the TimeXtender Repository database, and if so, which tables/views are recommended the TimeXtender API, and what permissions are required Why Admin → Security / Permissions is not visible in our portal, and wha

Hi All,My current employer uses a DWH with a lot of views (with unions and so on), which leads to slow execution packages. I would like to enhance the performance (rebuilding the views is one thing). But i woud also like to enhance the performance with indexing. So i have setup up one of our Dimension tables as seen below:Is our SupernaturalKey (SNK) Are our PK'sWhat would you guys recommend? Which type of indexing for SNK (nonclustered, unique index, Columnstore Index, Clusttered ColumnStore Index)?Which type of index do you recommend for PK's (nonclustered, unique index, Columnstore Index, Clusttered ColumnStore Index) ? Which type of indexing SNK and PK in fact tables?  

Hello, Three of my collegueas where developing in our dev enviroment. And when trying to save the project, we got the error:An error occurred while saving the project to the database.An item with the same key has already been added.Detailed error:An error occurred while saving the project to the database.Er is al een item met dezelfde sleutel toegevoegd.Details:Er is al een item met dezelfde sleutel toegevoegd.Module: mscorlibSystem.ArgumentExceptionbij System.ThrowHelper.ThrowArgumentException(ExceptionResource resource)bij System.Collections.Generic.Dictionary`2.Insert(TKey key, TValue value, Boolean add)bij TimeXtender.DataManager.ProjectPerspective.AddProjectPerspectiveItem(ProjectPerspectiveItemsRow itemRow)bij TimeXtender.DataManager.Project.InitProject(Int32 version, DsProject loadDsProject, DataSet adaptersDataSet, ProjectsRow projectRow)bij TimeXtender.DataManager.Project.LoadProject(DsProject _dsProject)bij TimeXtender.DataManager.Project.<>c__DisplayClass136_0.<Load

Hi All,In TX 20.10.45.64 we would like to work with Work Items to prefent Project Locks. There is no option to Release the work item and make it green. I already tried:Deploy Execut Save and Reload the projectWe only have the option to update, edit or delete the workitem.  How do we release these items and make them green? 

Hi team I was recently working on a table where I added a hash field and configured it for incremental loading. However, the initial deployment failed because the table was missing the required column for the incremental process. When I attempted to redeploy to fix it, I hit a roadblock: a duplicate key error in dbo.DataFields with the value (ea41fe97-e408-4504-9164-239dfcdee390, 148, 99999999).  

Hi,I have a customer working with TimeXtender 7150.1 connected to Snowflake.I have a .csv data-source connected through the Ingest. The ingest settings have a number of header rows removed. But in the data received, I have Totaal-rows that I want to skip or filter out.The option to “filter rows” by right-clicking the data-source in the Ingest area is not available for this source: My next option would be to add a data selection rule to the table connected to the ingestion. But that has no effect: my result table has two rows (one from each .csv source) with Totaal.The above data selection rule seems not to be used. The only thing I can do to get it working, as far as I am aware, is to manually add a line of code, using the “Custom Code” option:But I was told / had learned that using the data selection rule, would automatically change the transformation script.I am not a fan of this solution, as it is not immediately clear to a new user (or to me in a year time) that I have changed the

Hi team, I am encountering a timeout issue when moving large tables. The problem seems to be originating from the cleanup stored procedure (clean sp), as smaller tables are processed without incident. Could you please look into this issue?  

I’m migrating a large SQL Dedicated Pool data warehouse to Fabric Lakehouse.I’ve run into Fabric’s workspace limit of a thousand items.What’s the best-practices workaround?

Hi TX Support, We're encountering a strange issue. We've setup an deliver instance of the REST API endpoint type. The model contains a dimension table. In the MDW layer this table contains 7844 records. Now when we pull the data from the REST endpoint we get 7844 as well, however the content is incorrect. Instead of the 7844 individual records we get 6974 records, 870 of those are duplicated and consequently 870 other records are missing. Example:DIM_Location TableLocation, Code, DW_Id Original MDW data LocationA, 123, 1LocationB, 345, 2LocationC, 678, 3 REST dataLocationA, 123, 1LocationB, 345, 2LocationB, 345, 2Missing Record in the REST dataLocationC, 678, 3Now I did tried everything I could think of in terms of troubleshooting. Everything is deployed, executed, refreshed and running. Going through the data I made an important observation. The set that is duplicated all are records that reside on the first page. They are then duplicated across the remaining pages. See the screenshot

Hi all,I'm working in TimeXtender (TDI) and ran into something I can't quite figure out. When I create a new schema in my Prepare instance, I want to assign it to a table that already exists in that same instance. The schema does appear as an option when I try to link it to the table. However, it gives an error upon deployment (of that table), at least until after I deploy the instance. Is this expected behavior?Or am I perhaps creating them in the wrong order or place? 

Hello,I have a question about the RAW tables in TimeXtender. I noticed that the data types of the fields in the RAW table change when I modify them in the tool. I was under the impression that RAW tables should keep the original data types from the source, since this would be “RAW”.I ran into trouble when loading transferring from the ODX. I changed the datatype of a field and even added a transformation that resulted in a CAST in the transformation view. However, I got an error message saying: The given value of type String from the data source cannot be converted to type float of the specified target column.This should not be, because the source contains only data that is/should be convertible to a float. The Raw table has no data, which I find strange. Changing the fields data type to float has also changed the datatype of the Raw table. I would expect the Raw table to always have the same datatype as the source and that the changing of datatypes is handled in the transformation vie

I have an issue, and i have not been able to identify if this is a TX issue or a DWH issue.This is how TX is set up:I have a MS PowerApp that writes to the APP database.I can see that the records are in the DWH, like in the table below.  I have confirmed that the entetried are in APP-stack.At 07:30 in the morning, execution package runs, then this happens:Meaning, the DW_ID that were previously Bob’s and Marlen’s are now overwritten by Kim and Tom.There are no traced of DW_ID 101 and 103 being registrered by Bob and Marlen either in ODX or in DSA.I dont know if it happens before the execution job runs, during or after. I have been scratching my head for the past couple of weeks.

Hi community,I created a role in the MDW and added Grant permission on one specific database schema. This schema contains some tables and some Stored Procedures. This role need to be able to select data in the tables and execute the stored procedures. I created my Object Security Setup using these settings:  When I login using SSMS (using the user in the role), I see the correct tables and I'm able to select data in the tables. But I'm not able to see the stored procedures. When I check the permissions on the ‘ExcelAddin’ database schema. Only the ‘select’ permission in granted. Is it possible to grant permission for ‘execute’ as well? Using TimeXtender?

Hi,I have a client that uses TDI 7026.1 and i need to install it on another computer, but on the web I can only find 7050.1. Any possibility to find the earlier version somewhere?/Leo