Forum Discussion
Different Power BI instances generate different SQL codes!!
One more note - the report works perfectly on my Power BI Desktop and on client Power BI Desktop (December 2017)
It's only the online version that has this problem so it's probably connected to the online version or the gateway.
Both gateways are version 14.16.6XXX.X - they differ in the X numbers.
Hi RafalK,
When you publish the report to service, and refresh the data using gateway, it will return error because the datasource credentials in your service is different from client server. Let the client re-edit the credential of dataset, and make it work fine.
Best Regard,
Angelia
We are changing the credensials and HANA server adress each time we republish.
Validating the code changes nothing. We will play with it a little to see if it helps.
The problem with the SQL generated is that it is
1. different depanding on the enviroment
2. there is an error in one of those SQL codes that is pushed to HANA:
here is the part of the code:
SELECT ...
FROM (...) AS "basetable0"
INNER JOIN (
(SELECT 'XXXXYYYY' AS "c118")) AS "semijoin1"ON
...
The code is missing "From".
If i write it like this:
SELECT ...
FROM (...) AS "basetable0"
INNER JOIN (
(SELECT 'XXXXYYYY' AS "c118" FROM dummy)) AS "semijoin1"ON
...
it works. I am making the change to the code in SAP HANA Studio as I can't change the code that Power BI changes automaticaly.
This part of the code is not an SQL that I pass on in Query Editor but rather something that Power BI adds when I filter one visual by another.
In the OTHER SQL that is actually working this filter is applied with a simple WHERE:
SELECT ...
FROM (...) AS "t10"
WHERE ("t10"."PROJECT" = 'XXXXYYYY')
...
