So, I'm pretty much a noob at WebFOCUS but I have a question on some of the internal workings of the product.

I have a FEX that is producing a rather large PDF file (22 MB / 618 pages) and I notice the first run takes several minutes but the subsequent run(s) (for the same parameter information) take a considerably shorter time. This leads me to believe there is some caching occuring on the server. FWIW, the FEX is using ON TABLE HOLD statements. I'm guessing this is the case, but would someone mind confirming? Thanks!

- ABT

P.S. There is no burning urgency or break/fix associated with this question, - just something to satisfy my curiosity.

This message has been edited. Last edited by: ABT,

I doubt there's any WebFOCUS "caching" occurring. Rather, it's probably the intelligent DBMS software that determines that the request is the same as a recent one...

Depending on where you're running your request from, it is also possible that the caching is taking place in your own Internet browser.

For instance, most of our reports are accessed through an ASP-driven web application with final calls to the WFServlet and whenever we submitted a URL that looked exactly the same as that of another previously sent request ( from the same workstation) , the results came back almost immediately and we could tell they represented a "cached" version because they didn't reflect last minute changes to either the report layout or the data.

My workaround was to make sure (as much as I can) that no URL matches another one sent from the same workstation during a short time span, so I just use a dummy parameter with a looooong random number (generated with Javascript) that always gets added to the URL before WF receives the request.

Indeed, the caching is done by the browser. There are ways to specify that whatever is being sent to the browser expires immediately. Effectively, the browser will then never use caching. I just forgot where this can be set - could be some browser setting, but I think it is more likely a webserver setting.

Indeed!

This would not be a browser caching issue, unless the browser was caching the data returned by the database calls.

From the first query to the second, I am changing the internal structure of the program -- nothing that will affect the data returned. Pre change, it takes a few minutes. Post change it returns in considerably less time.

Best answer so far is Francis'.

- ABT

quote:

Best answer so far is Francis'.

Sorry, I did not know this was a contest. However I'm with you on that! Francis is a great supportive guy in the forum and his tips are excellent not to mention timely.

Please let us know once you find the culprit in your issue so we can keep that information handy in case we face similar cases in our own environments.

Turn on SQL tracing. If the SQL looks the same for both requests, it's probably the DBMS (of which you've told us nothing about).

Thanks everyone for your thoughts. This was not a contest and not an issue, but I appreciate all of your input.