Row Duplicate from Int16 overflow

[JoshKushner]

Has anyone experienced a row duplicating 65,535x in their output to excel?

The canvas pushes the right number of rows to the output tool, but the output tool seems to hit an Int16 overflow duplicating some rows 2^16x.

Thanks to @Claje for identifying this was an Int16 overflow issue.

[danilang]

Hi @JoshKushner 

Can you provide a link to the original post?

Dan

[JoshKushner]

Hi Dan,

Could you please clarify? This should be the first thread with this topic.

If this is regarding a link to the original workflow, that unfortunately can't be provided.

Sorry this is pretty unhelpful.


After testing, it seems like a highly unique issue. At this point I'm most interested if anyone else has seen this specifically happen to them.

Bug specifics:
-exclusively occured with a file name, data combination
-exclusively occured writing to a .xlsx file (not legacy)

[danilang]

Hi @JoshKushner 

I thought that since you mentioned @Claje, that he had helped you on an original community post.  I assume that he also went through the obvious points about record count, etc.  

Can you explain a bit more.  Do you have a case where you have X records in Alteryx at the output but X*65536 when you open it in excel?

Dan

[JoshKushner]

Alteryx has 1199 rows.

Two of those rows duplicate 2^16x.

This is unique as a result of a specific workflow. Can't be duplicated by
copying the data from one file to another. (Could be a metadata issue)

[danilang]

@JoshKushner 

Does the duplication happen in the output tool, or is the explosion caused by another tool?  Is there ever a point in workflow where you see the 2^16 duplicates?  

Dan

[JoshKushner]

Duplication occurs in the Output tool.
I have confirmed this is a metadata issue. I removed all columns (and other combos in-between) and it still overflowed.

I'm worried this is a Python tool issue.

[JoshKushner · Accepted solution]

After extensive testing of this issue, the row duplication originated from a single duplicate row cascaded day-over-day, joined twice in each iteration.(256*256=2^16).

Closing this out confirming it is not an Alteryx issue.