Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi guys,
we use the QlikSense Server at one of our client with Windows Server 2008 Operations System.
QlikSense version is 3.1 RS3. ODBC driver version: Open Edge 10.1B (32 bit)
I connect to the database via ODBC.
A very simple script with few records works:
Here is the script:
LOAD "pt_part",
"pt_um";
SQL SELECT TOP 8000
"pt_part",
"pt_um"
FROM PUB."pt_mstr";
Here is the result:
Started loading data
Connecting to TEKA-QADDB (mofemws_mfg) Connected pt_mstr Lines fetched: 8 000
Creating search index Search index creation completed successfully
App saved
Finished successfully
0 forced error(s)
0 synthetic key(s)
But above 9000 records the script does not work any more and the number of lines fetched is zero.
Here is the script:
LOAD "pt_part",
"pt_um";
SQL SELECT TOP 9000
"pt_part",
"pt_um"
FROM PUB."pt_mstr";
Here is the result:
Started loading data
Connecting to TEKA-QADDB (mofemws_mfg) Connected
pt_mstr Lines fetched: 0 Creating search index Search index creation completed successfully
App saved
Finished successfully
0 forced error(s)
0 synthetic key(s)
I tried to load other tables, but the results were the same.
However it worked just fine using Qlik Sense Desktop software on my own computer (Windows 7 OS) via VPN connection.
I checked the followings:
Does anybody have any idea?
Thank you in advance!
David Barcza
Compare both DSN configuration. I doubt they are not identical.
What you mean DSN configuration? Is it about the Data Source Connection? Because the ODBC settings are good I have already checked it. Anyway I tried to change the Fetch Array Size too but no different.
OK guys, it seems so I found the source of the problem. Probably the problem was with the ODBC settings. The ODBC fetch array size was 5000 and the network connection to the DB server is so poor that the ODBC driver could not handle this size of data package transfer.
So after we decreased the fetch array size to 50 the script works agains. The speed of the loading data process is not so good, but it works.