Mismatched Column Definitions Error

[LoganMayer1994]

Hello,

I am trying to utilize the Tableau Output tool and I am running into an error that I have not been able to find a resolution for. I have used the current Tableau Output tool (I have the most recent version of Alteryx) and even the previous Tableau Publish tool so I am quite familiar with setting up connections to publish to Tableau. I am updating a workflow from Tableau Publish to Tableau Output but I am running into a TypeError: mismatched column definitions. I am using the same Alteryx database source and publishing to the same Tableau data source. Can anyone help me to understand why I am getting this error?

[LindonB]

I've experienced a similar error in the past trying to write to a DB table when the class of a field changed. It appears that the problem is the "Accounting Doc Num" field, which is the type of field that could be number or a string (and it seems that it is a Type = TEXT). Not sure if this is it, but I'd first try casting it to a number (or string, if that field is currently a number). 

[LoganMayer1994]

That was my first thought as well too but it is giving the same error across the board for each column. For example I have a column that is just numbers and the error says "Type=Double" then the Nullability piece. I input a select tool to check/change the data type and it is in fact a double already. All the items with "Type=TEXT" are also V-string. So, I am just at a loss on what I need to adjust to get the Tableau Output tool to publish.

[srachelle]

Did you ever get a resolution on this issue? We're having the same since updating to the tool version "Tableau_py3.8_v1.4.2.yxi"

We're on version: 2023.2.1.226 Patch: 6

[Darren_Hutson]

I had this same issue and was able to resolve it by using the select tool and changing any field with the type "byte" to double and it seem to publish fine after that.

[eric_strein]

Has anyone resolved this issue? We too are having the same error since updating the Tableau output tool to version 1.4.2. I have multiple workflows with this output tool but only one is encountering this error.

I've reviewed the Tableau data source types and there is no actual type mismatch, all data types are correct on both sides (Alteryx and Tableau).

We're using version 2024.1.1.136 Patch: 4

Thanks!

[Darren_Hutson]

Are any of your field types in Alteryx labeled "byte"? Even though they previously processed fine into Tableau (both systems would say the type was "byte"), the new tool no longer handled that field type properly and I had to change it to Double in the select tool.

[amruta · Accepted solution]

I upgraded the Alteryx server from 2023.2 to 2024.1 

changed to use the Tableau output to version Tableau_py3.10_v1.4.2.yxi.

We have been experiencing the same issue with some workflows which were previously running successfully with no issues in older version.

we kind of figured out the cause. First, copy and paste the error message that starts with Mismatched column definitions: in notepad++ and re-arrange the text so it is readable. You will clearly notice that each column is listed twice and has also data type listed. Start comparing the data types of each column.  I used Tools like "Compare it" to do comparison of columns if you have 50+columns like I did

You will notice at least one column where the data type does not match, that is the field that is causing issue with the upload. At least now we have narrowed down the issue.

We noticed in our case, its mostly DATE column that had NULL values is the cause of the problem. (so not all Date columns will cause the issue, only those where there are null values)

NOTE that even though it is DATE column, in the error message it will call it as data type DOUBLE (no clue why)

FIX:

(1) change the data type to V_STRING (instead of date) for the column that is affected (in our case the "Created Date") (refer screenshots)

(2) most time the replacing the existing Tableau extract will not work, you will have to create a completely new output-name. (and then do a replace data source in Tableau desktop) 

I hope this helps someone facing this issue with Tableau output

[amruta]

AGREE!

Byte is not supported anymore > needed that to be changed to Int16

[jessarthur]

@amruta your solution was immensely helpful. We had the same error message where columns were appearing twice in the error message. We did a side-by-side comparison and noticed one of our Date columns was showing as date in one list and double in the other. We changed the date to a string as you suggested and re-ran the workflow with no issues. Thank you!