TimeXtender Desktop Q&A

Previous Paragraph. This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.  This one is super long though so you can see the difference when paragraphs span multiple line breaks.Heading OneNext Paragraph. This one is super long though so you can see the difference when paragraphs span multiple line breaks. Next Paragraph. This

Hi all,  I need to create an Identity column, How could I create it in TX? Thanks Ignacio

Hi, Let's assume I am importing from CSV file. The Data is very dirty and somehow I have stings in my int filed "ID" so the filed is interpreted as NVARCHAR(MAX). If I change data type of the field (right click > Edit Field) from string back to integer I receive a runtime error during the execution. What I was expecting was to see was corresponding entries in the _M and _L tables. Is this not part of data cleansing after all? We have this problem all over and could not find a better solution than to build custom views with TRY_CAST etc. where we lose lineage etc. What is the best way to clean up such data (keep valid rows, write errors/warnings on invalid rows)? We are talking 100+ tables so I'm looking for a highly scalable solution here not a one time work around. Thanks and BR, Tobias

After trying to import data from a REST source in the ODX I got an SQL Exception with error code 4815. After a bit of Google I ended up on this page, telling me that it had something to do with SQLBulk copy not able to match the columns in source and destination or the value of a column was larger than in the destination allowed (e.g. string with length of 200 -> nvarchar(100)). Because TimeXtender defines the columns it probably didn't have something to do with that. Because the RSD I created didn't have a column size specified, all my string type field allowed 2000 chars. Thinking that none of the string fields were going to be larger, I started changing int to double and after that all the fields to string and later even to text. None of this worked which send me back to start. After properly examining the parquet file I found 1 record that had a text field that exceeded the 2000 char limit. The next time I get this error I will do the following steps: 1. Examine data type of API

Hi,  I'd love to get your thoughts on issue we're dealing in the past few days. We've noticed that transfer time from ODX to DSA is growing rapidly. This as significant impact on total running time of DWH and SLA towards our client Execution log of DSA  As you can see. as insert and cleansing tasks remains pretty much the same, transfer time is growing. In terms of volume, data hasn't grown significantly   Batch Id 13428 (first) - 46347863 rows Batch Id 13481 (last)  46536187 rows (+0.4%) difference in transfer time is more apparent as you zoom in into the package execution (Gantt chart) Batch Id 13428   vs last batch (13481) There is no inconsistencies in ODX que or ODX service logs found.  I've also checked it there is long running transactions (sp_who2) in both TX and DWH server but could find anything out of the ordinary    Our setup consists of ODX repository and DWH storage on same Azure Managed Instance. TX application server is (still) on-prem  My burning question is: what c

Would be great if within TX (or via external script) you could trigger Data Source transfer tasks as an initial Step within an Execution Package (or batch), so that once the transfer of new data is complete, the execution package runs the next dependant step?  This way, if the transfer task from the data source to the DL takes longer (more data than expected) then the execution package won't run until it completes.  There seems to be no method currently to trigger a transfer load within the execution packages - and therefore no way to make the two different scheduling tools (it's really that) align with each other intelligently.

Does anyone know whether or not the rules you put in the over ride data types area are applied sequentially like they would be in a SQL case statement (i.e. once a field evaluates to TRUE it ignores the subsequent logic)?

I am new to TX and data lakes.  I have a common problem that others have commented on and would like to know if: 1) The functionality now exists, or if it does not 2) How are others getting around the issue? The issue is that I need to be able to segregate data sources in the data lake.  I am pulling in data from multiple SQL sources where the source schema and table names are the same, and I need to differentiate those tables in the data lake.  By schema would be best, but if that is not possible, I need some other ideas.  Thanks!

Hi, I am a bit unclear on how to configure Max Threads for an execution package. The way I perceive it now is: if I run on 4 cores, I choose 4 Max Threads, if I run on 16 cores, I choose 16 Max Threads. i.e. N Cores -> N Max Threads. Is this the correct approach or are there other factors that need to be accounted for? 

Hello, I have a problem that i am struggling with. I have a table that contains field called [Period] and it stores month dates in following int format: 2201, 2202, 2203 which i need to translate to 2022-01-01, 2022-02-01, 2022-03-01 etc. to date format (always 1st day of month) The problem is i can convert the source format to date format in the view, but when i am trying to generate a table from that view, i always get an error that [Period] cannot be converted to date format. I have different options and nothing has worked as timextender always fails to convert to date format. Everything that failed on timextender, worked perfectly on MSSQL so i am not sure what is wrong. Thanks for your help

Hi, I'm searching for a way to get a table insert incremental. I have a table (about 30 million records). This has a row for each hour of the day for 10.000 machines, so this table is getting bigger very quickly. I do a lot of transformations in this table. This resulted in a load time of over 10 hours. I've discussed this with our TimeXtender Solution Specialist and we've decided to split the table into multiple smaller tables.  Now I do have 1 table (loaded incrementally from ODX), this table has a lot of lookups and simple transformations. This table is loading quickly because it's incrementally loaded. The result of this table is inserted in a new table (Raw table), using a table insert. This table is converting cumulative numbers to noncumulative, using the SQL functions LAG, OVER, PARTITION BY and ORDER BY. Because it is a table insert I have no idea how to do this incremental. I've set NonClusted Indexes (also on Raw table) on all fields used in the PARTION BY function. And enab

When we perform a transfer from the acceptance environment to the production environment, there are settings that you do not want to take with you. First the time schedule. We have an Execution package for both acceptance and production. These should run at different times.  Secondly, the acceptance environment naturally refers to other TimeXtender databases than the production environment. Nevertheless, the database links are taken literally after the transfer.  Third, when you accept a change in acceptance, you transfer to production. However, when you have executed these, the sources are literally copied from the acceptance. We have circumvented the latter by allowing acceptance to refer to production, but this is not always desirable. Fourth, it would be nice to be able to include project perspectives during the transfer. It would be nice if you could set the above points as settings to transfer. 

Is it possible to or are there any limitation on using composite models in PowerBI in unison with the Semantic model created on TX? Regards Mads 

Hi Is it possible to use DAX Studio og Tabular Editor on Semantic models?Regards Mads

Never had this before ...Trying to execute just the object I am working on..It auto selects the whole project ...where to change that setting...Very strange... 20.10.25.64

When I do 1 on the exercise, I get a pop-up information message "No constraint suggestions are available" Right click DSA node and choose Automate -> Add Suggested Constraints Don't know how to resolve this?

Hi all, I'll try to find out how to get rows in the _L table when I use a field validation.   For field validation I can decide if I get a warning or an error information. But I only get a message in the _M like this: How can I get the rows that don't match the field validation criterion to get an entry in the _L table? 

Hi, we are receiving a file where the order of the rows is important information that needs to be maintained for our transformation process. When TimeXtender loads a text file, are the rows loaded in order? (I.e. will the DW_Id in the Business Unit table act as row numbers of the source table). Thanks, Mark

Timextender rearrange include steps in Execution Package 1) yes i know you can hold [Alt] and use Arrowkeys up and down but really drag and drop should be possible   Also make the "all Steps" "include Steps" and "Exclude Steps" 3 separate columns and the click options at the bottom - make it better 1) post execution first 2) during Execution  middle 3) after exe last like we use it ...

Hi, Currently we have execution packages set to retry upon failure, this is set as Retries = 5 , Retries per step = 3Occasionally we encounter a failure, however it seems we can then only access the logs for the most recent fail, is there a way to get the logs for all of the failures?Similarly, is there a way so get logs of a failed execution step when the full execution package succeeded after a retry?

When I have two tables, how can i make a left outer join between 2 tables. Or do i always need to use a custom view for this situation.  Table 1 = Adressid = 1id = 2 Table 2 = Relationpersonid = 1personid = 2 Table 3 id = 1, personid = 1, personid = 2id 2 , null          

Hi all, We recently started developing a new data platform using TimeXtender with Tableau as a visualization tool. We are currently determining how to setup our endpoint in TimeXtender for Tableau as we have several options, each with their pro's and con's. Im wondering what choice others made and if maybe there's something we haven't thought about yet while trying to make this choice. As we see it now we have three main options: - Multiple .tds files, one for each fact table. As Tableau requires all tables to have a relation with eachother it is not possible to have a .tds file endpoint with all facts as not all facts relate to all dimensions. Pro's: smaller, efficient sources with good manageable security. Cons: Can become difficult to maintain with lots of .tds which share dimensions, for example when an attribute is added to a dimension this has to be done multiple times. When using multiple .tds sources in a Tableau sheet there's no way to link shared dimensions and thus fltering

It would be really beneficial if functionality was added to the Semantic layer, such that SSAS Tabular partitions could be defined within the tool.Ideally, one would be able to right click on a table within the semantic model and choose partitioning, which would allow you to pick a field, like a Date field and configure it so that partitions on say Year, or Quarter, or Month, or Week would be dynamically created with each deployment of the tabular model. Additionally, it would also be great if once TX had visibility of the metadata of the partitions it has created, it would be extremely beneficial if we could then decide which partitions are executed in a given execution package.A little background for this request, we are in the middle of a major AX and data warehouse project, and we are building out quite a few SSAS tabular models to be consumed by Power BI.  Unfortunately, with the current setup of the semantic layer, we are having to consider managing tabular models outside of Time

We would like to be able to set specific schedules for our different environments. For example som execution packages could run on schedule in our QA environment during testing. And the TXDictionary should run on schedule in all environments.

It would be great if in the Multi Environment Transfer Step it would be possible to refresh the Data source as well.Currently I have to add new tables in Dev en Prod environment on a weekly basis (chaning ERP systems).It happens that a table is added in Dev, but that table is not yet avialable in Prod. So when I transfer the project and deploy I do not get a warning that the table does not exists. So after deployment, the schedule will run in morning and it will fail on the missing table.Only after a refresh of the Data source the project recognises the missing table and gives a warning. Now I have to in Dev transfer project and deploy to Prod and then open the Prod environment itself to refresh data source in order to check if all tables are available or not.