Free Trial

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.

Data Loss using data-stream Out in-db tool

EnaK
7 - Meteor

I have an issue with data stream out in-db tool and was wondering if anyone has a solution:

 

it is randomly losing data, example below shows same filters applied but different output as a result before and after data stream out used: 

Data stream-out data loss.jpg

 

 

 

30 REPLIES 30
EnaK
7 - Meteor

1) amp is off, 1st thing I did 

2) HANA

 

table is not changing for sure, i am running simple query in parallel and getting expected results

EnaK
7 - Meteor

no filters... format is the same.. nothing different

HANA is the database behind

Data stream-out data loss.jpg

EnaK
7 - Meteor

no, absolutely not :(

EnaK
7 - Meteor

I keep responding, but my response won't show up for some reason

1) 1st thing i did was turn off amp

2) HANA

data is not changing, i run query in parallel and see correct results

EnaK
7 - Meteor

no filters applied, please see the difference in results below:

Data stream-out data loss.jpg

apathetichell
19 - Altair

add a select tool before summarize tool on canvas - as mentioned sometimes things misconvert. My hope is you see a fixed decimal ending in 0 meaning all your values are being rounded.

 

also - can you add a count to both summarize tools?

EnaK
7 - Meteor

I do need to see decimals though as in original data set, I want to keep the numbers as they are

apathetichell
19 - Altair

Here's what I've seen - let's say you have 10mm records in your dataset. you have lots of amounts like .25, .23 etc. these become 0 when fixed decimal creates a truncated value. If you can add a select before the summarize tool and post the results - that would be helpful.

EnaK
7 - Meteor

Data stream-out data loss.jpg

apathetichell
19 - Altair

Ok! So that's not the issue - I guess a count? Not really sure if there's something HANA specific re: data truncation via the ODBC. Maybe someone else knows about it. 

 

My next suggestion would be to look for line item differences and seeing what have in common.

Labels
Top Solution Authors