Qlik Community

QlikView Scripting

Discussion Board for collaboration on QlikView Scripting.

Highlighted
Not applicable

Reading SAS datasets located on Unix-server

Hi,

I just got involved in a project where we have to read a lot of SAS datasets for our QlikView solution. The SAS-data is located on a Unix Solaris 64-bit server. On another machine we have the QlikView-server. I am trying to figure out how to read the data directly from QV. The only information I have found involves installing an ODBC driver for SAS on the QlikView-server as well as installing a local SAS server on the QV-machine. This is the document someone at QlikTech gave me regarding this: http://community.qlik.com/servlet/JiveServlet/downloadBody/1998-102-1-1993/How%20to%20Setup%20ODBC%2....

However, we are not allowed to install a SAS server on the QV-server. And the document seems based on the assumption that the SAS server is Windows based and not Unix.

I am not at all experienced in matters involving connectivity between different environments, so any advice would be appreciated. How can I read SAS datasets from an Unix server directly from QlikView installed on another server?

Thanks,

David

Tags (4)
18 Replies
MVP & Luminary
MVP & Luminary

Re: Reading SAS datasets located on Unix-server

Hi David,

I'm not sure and haven't tested so far but maybe a JDBC connect could help:

http://support.sas.com/documentation/onlinedoc/jdbc/

- Ralf

Not applicable

Re: Reading SAS datasets located on Unix-server

Hi Ralf,

Thank you for the tip. I've also read other threads on this forum where you discuss the QlikView JDBC Connector. And also checked out the product at http://www.tiq-solutions.de/display/enghome/ENJDBC.

As I mentioned above, I'm not at all knowledgeable in the area of connectivity, and I wont be the one doing any installation on either our SAS-server or our QV-server. I'm merely trying to find information I can forward to the people responsible for this.

How would it work? Is it correctly understood by me that the SAS JDBC driver should be installed on the SAS-server and the QlikView JDCB Connector installed on the QV-server? But as SAS is not listed as one of the possible data sources for the JDBC Connector, what are the chances that it would work? Why wouldn't it work?

I really appreciate any advice you can give me.

Sincerely,

David 

MVP & Luminary
MVP & Luminary

Re: Reading SAS datasets located on Unix-server

Hi David,

we haven't tried SAS as JDBC source so far. There are hundreds of possible JDBC sources (or even more.. ;-). We can not test all of them upfront because this would be and endless venture. But, we will test on a clients request base and I see no problem since JDBC in fact is a standard and vendors have to follow the implementation rules. Of course there could be some obstacles but we can adapt the JDBC connector to a workable solution. Which we already did in some cases because of some inadequate vendor implementations.

- Ralf

PS: to answer your second question the JDBC connector must be installed on the QlikView server. The SAS JDBC driver must be configured on JDBC connector (on the QlikView server). I guess it's just an .jar file..

Not applicable

Re: Reading SAS datasets located on Unix-server

ok, great Ralf. Thank you. I will forward the information that you have provided to my colleagues. Hopefully we can test the solution you suggest and I will get back here and tell you how whether or nor we got it to work.

Cheers,

David

Not applicable

Re: Reading SAS datasets located on Unix-server

Hi again Ralf,

After a lot of internal political discussion regarding whom in our organisation would be responsible for making QlikView query SAS we finally got going according to your suggestion.

So we have the SAS-provided JDBC driver installed. But when we try to install the JDBC QlikView Connector it asks us to enable the Internet connection. This is something we can't do. My client is a bank and that server is not allowed to be connected to the internet. Do you have suggestions?

Thank you,

David

JDBC Connector.bmp

MVP & Luminary
MVP & Luminary

Re: Reading SAS datasets located on Unix-server

Hi David,

please see attached file for offline activation.

- Ralf

Not applicable

Re: Reading SAS datasets located on Unix-server

Hi again Ralf,

My IT-department still haven't been able to get it to work. I believe they've been in contact with you or at least with someone at TIQ Solutions. I've been kind of out of the loop focusing on building the parts of the QV-applications that do not need SAS data. However, now I am at a point where I really need to be able to read our SAS data sets. But internally at my client this issue has turned into a blame game between the QlikView, Windows, Unix and SAS administrators and nothing gets done. I've been playing around on my own computer over the weekend to see if I can get it to work. 

I've installed the latest JDBC driver from SAS and the Java environment. I have my local QV-client and a trial version of the JDBC QlikView Connector. I've tried loads of different configurations but whenever I try to connect by providing the connection URL, I get the message "No suitable driver found".

Can you assist me in what I am doing wrong? And is there any kind of documentation for your product?

Attached are screendumps of my latest configuration of the JDBC QlikView Connector.

Thanks again,

David

JDBC1.bmpJDBC2.bmp

JDBC3.bmp

MVP & Luminary
MVP & Luminary

Re: Reading SAS datasets located on Unix-server

Hi David,

please get in touch with me directly: ralf.becher@tiq-solutions.de

- Ralf

Not applicable

Re: Reading SAS datasets located on Unix-server

Hello,

It's been a while since I started this thread, and finally we have got it working. It took a lot of trial and error and a lot of help from Ralf B. but we now read SAS data sets from Qlikview using QlikView JDBC Connector.

I have a lot of information gained from this experience that I would like to share here on the forum. But at the moment I am overloaded with work trying to meet a deadline of last of June. But after that, early July I will write another post in detail describing what we had to do, and what we did that didn't work.

//David