Skip to main content
Announcements
Defect acknowledgement with Nprinting Engine May 2022 SR2, please READ HERE
cancel
Showing results for 
Search instead for 
Did you mean: 
kswindlehurst
Contributor III
Contributor III

Why does Qlik insist on calling it metadata?

Yeah, I can understand the need to refresh metadata if the source design has changed, and if you need to schedule data loads from the source because you can't manage it on-demand, that's fine.   But don't start conflating the two concepts.    Metadata is descriptive; data is content.  Single-version-of-the-truth should apply to terminology as well.

Labels (2)
8 Replies
Lech_Miszkiewicz
Partner Ambassador/MVP
Partner Ambassador/MVP

Hi @kswindlehurst 

Is there partiqual question in your post? do you need any clarification? or... did you just want to went little bit (which is fine with me 😁😉) and wanted to let everyone know(which is also fine with me 😁😉) ?

What I mean is if there is any particular scenario you want to talk about or discus it will be important to choose correct version of NPrinting to start with as well as tell us about a setup (QlikView - which exact version). As you know it is quite important to establish those basics  if we want to chat about the rest. 

cheers

Lech

cheers Lech, When applicable please mark the correct/appropriate replies as "solution" (you can mark up to 3 "solutions". Please LIKE threads if the provided solution is helpful to the problem.
Ruggero_Piccoli
Support
Support

Hi,

"Metadata is descriptive" and what Qlik NPrinting loads in the Repository during a connection metadata reload is the description of the content of that connection which include the name of fields, their first 1,000 distinct values of each field, the objects (like charts, listbox, sheets, etc) in the source app/document. So when the content change, for example because a new chart is added, it is needed to reload the connection metadata to include it in the list you see when editing a template. 

Qlik NPrinting doesn't load all the data of a connected app.

Best Regards,

Ruggero



Best Regards,
Ruggero
---------------------------------------------
When applicable please mark the appropriate replies as CORRECT. This will help community members and Qlik Employees know which discussions have already been addressed and have a possible known solution. Please mark threads with a LIKE if the provided solution is helpful to the problem, but does not necessarily solve the indicated problem. You can mark multiple threads with LIKEs if you feel additional info is useful to others.
kswindlehurst
Contributor III
Contributor III
Author

Ruggero,

My own experience and this post appear to be in conflict with your description:

https://community.qlik.com/t5/Knowledge/NPrinting-reports-include-outdated-data-until-connection/ta-...

I would have no issue with using the term Metadata as you describe, but it appears to include more than that.  Perhaps that was the original intention but the situation changed?

-Kevin

 

 

kswindlehurst
Contributor III
Contributor III
Author

Lech,

I think there is a problem with the documentation and it's causing confusion.   Unless someone has a different interpretation of "Metadata", which I'm open to hearing.  Or perhaps the old data issue is limited to "Local" qlikview connections:  https://community.qlik.com/t5/Knowledge/NPrinting-reports-have-old-stale-data-with-local-NPrinting/t...

Yeah, I guess from this old post of yours(!) that the metadata refresh has an "off-label" use that fixes local connection problems:  https://community.qlik.com/t5/Qlik-NPrinting-Discussions/Nprinting-doesn-t-refresh-the-data/m-p/1137...

So I guess that's the deficiency in the documentation, not the use of "Metadata".  

Lech_Miszkiewicz
Partner Ambassador/MVP
Partner Ambassador/MVP

Hi Kevin

I see what you mean. In terms of clling nprinting cache a "metadata" I am still fine with it as it is actually as Ruggero mentioned in this scenario a "metadata".

 

The fact that we sometimes see old data in the reports and "generate metadata" reload must be used is for different reasons. I am not a Qlik R&D so I am not 100% sure but here is what I have found so far:

  1. QlikView local connection - When NPrinitng engine generates report it opens up local QV.exe instance and opens qvw in it. Since it is a localy opened file it will not get updated until NPrinting closes qv.exe and reopens it again to load the new version of qvw (new version being the one which was recently reloaded). NPrinting Engine does have this stupid behaviour of keeping some of the qv.exe instances open which is creating the problem. Triggering "generate metadata" closes all qv.exe instances and therefore allows for new qv.exe to be loaded. This is how i see this problem with LOCAL connection to QlikView. 
  2. There is also another instance when sometimes we observed issues with Qlik Sense variables not being immidiatelly updated and where obviously connection metadata reload would do almost the same as described in point above for LOCAL connection but in this case i guess it will reset "webrenderer" which produces Qlik Sense. I am chasing up some clarification under this post on the scenario 2: reports. https://community.qlik.com/t5/Qlik-NPrinting-Discussions/Task-does-not-refresh-filter-variable-prope...

Conclusion based on my understanding:

  • Metadata should be still called metadata in NPrinting
  • We probably should choose our words smarter and not say that reload of metadata is required to "bring new data to NPrinitng", instead we should probably say that reload of metadata connection is the process which disconnects instances of Qv.exe (LOCAL) and WebRenderer (QLIK SENSE) which results in the latest instance of the app to be connected. It is almost side effect of actual "metadata reload" process.

hope this helps

cheers Lech, When applicable please mark the correct/appropriate replies as "solution" (you can mark up to 3 "solutions". Please LIKE threads if the provided solution is helpful to the problem.
Ruggero_Piccoli
Support
Support

Hi @Lech_Miszkiewicz ,

1 it is not "stupid behaviour" is it "optimization" 😛 . The reason why the Engine keeps a Qv.exe process opened for each recently used QVW is that opening a QVW requires time. So the Engine suppose that if you used a QVW to generate reports there is a good probability that you will use the same next time and it keep it opened. The consequence is that a Qv.exe process is keep running. With small QVWs this optimization has not a green impac but we had cases where customers uses huge QVWs that require a long time to be opened. Instead when hardware resources are needed the Engine will kill the old Qv.exe process if the are keeping opened QVWs not used to generate the requested reports.

Yes, I agree, in those cases it is better to use a QlikView server. 

Best Regards,

Ruggero



Best Regards,
Ruggero
---------------------------------------------
When applicable please mark the appropriate replies as CORRECT. This will help community members and Qlik Employees know which discussions have already been addressed and have a possible known solution. Please mark threads with a LIKE if the provided solution is helpful to the problem, but does not necessarily solve the indicated problem. You can mark multiple threads with LIKEs if you feel additional info is useful to others.
Lech_Miszkiewicz
Partner Ambassador/MVP
Partner Ambassador/MVP

Yeah. I know the reason and agree with the optimisation aspect. Sorry about my bad choice of words 😬


I guess it would be good to have some setting which we as end users could choose if the connection should be closed each time or kept open as it is at the moment. 
There is no one right scenario in this case i guess. 

cheers Lech, When applicable please mark the correct/appropriate replies as "solution" (you can mark up to 3 "solutions". Please LIKE threads if the provided solution is helpful to the problem.
Ruggero_Piccoli
Support
Support

Don't worry @Lech_Miszkiewicz I understood your meaning :-).

I shared the info for all readers and I agree with the other parts of your answer. Thanks



Best Regards,
Ruggero
---------------------------------------------
When applicable please mark the appropriate replies as CORRECT. This will help community members and Qlik Employees know which discussions have already been addressed and have a possible known solution. Please mark threads with a LIKE if the provided solution is helpful to the problem, but does not necessarily solve the indicated problem. You can mark multiple threads with LIKEs if you feel additional info is useful to others.