Skip to main content
Announcements
Have questions about Qlik Connect? Join us live on April 10th, at 11 AM ET: SIGN UP NOW
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Missing data on reload

Hi All

I was wondering if any of you have had the same problem,

Am connecting to MS SQL 2005 using OLEDB 32 bit connection with QV10 SR3 and at random time there are chunks of missing data.

As you can imagine this is really compromises all my reports where it's got to the point am asking can i trust what

qlikvew is producing.

If any one else has had this probem and can give me some insight as to why this happens please do.

28 Replies
Not applicable
Author

Many thanks for your replies folks.

Angus - We don't feel that the issue was related to the locking and blocking as suggested before as nothing in our testing pointed us to this being a SQL Server -related issue only. 

Rob and Tom: our Support team currently employes a manual process for collating the list of people who have sumbitted a bug and cross references that with our internal customer databases to build the list of people to whom to send an email to. It's possible in your instances that you were missed in error as part of that manual process and I will reach out to our Support team and investigate. Going forward our goal is to have a much more robust automated system in place.

Many thanks again,

John

Not applicable
Author

I, too, am seeing this issue.  My load is from SQL server and one table is 90 million rows, selected with no where clause.  On my desktop, where I do all of my development, each load will only return between 20 and 30 million records.  When I open the document in QV11 on the server and perform the reload, the correct amount of records are returned. 

So...same document on two different computers runs differently.  The only difference is OS - both 64 bit, but Windows 7 on the desktop and Windows Server 2008 R2 Standard on the server.

This is with QV11.  Has the bug been addressed in 11?

Not applicable
Author

Our particular bug (number 47298) is available as a patch in V11 SR1 and is now included in the latest update V11 SR2.  This bug caused intermittent data loss from random tables on the SAME computer.  Your issue sounds unrelated and may be due to differences in memory between your desktop and the server.   Good luck

Not applicable
Author

Mayday! Mayday!

I upgraded to V11SR2 and I am still getting the incorrect number of records retrieved.  In fact, this time it retrieved FEWER!

And now...the last reload on the server also missed records.  Not nearly as many, but I fear this may be the beginning of something bad.

Anonymous
Not applicable
Author

We are seeing this as well, intermittently but more frequent with QV11SR2 installed. Will be curious to see if the community has ideas- we are going to start Oracle traces on the offending loads starting next week, to rule out database-side issues (middle-of-the-night scans of a transactional database, should be limited traffic).

We see corruption of our incremental loaded tables as well, almost like the incremental logic garbles the key values on occasion.

Good luck, will share anything valuable that we find.

Cheers,
DK

cjzimm76
Partner - Creator
Partner - Creator

We are using V11 initial release and are having the same issue.  I was wondering if anyone has an update on this issue since October.

Anonymous
Not applicable
Author

We ran traces for a couple of weeks and didn't find anything valuable in them. The issue is still intermittent for us, however it does seem to be related to only one of our source databases (which is under a lesser load than others which we capture from). We had a small amount of success with putting a Sleep step into the incremental flow, but it did hasn't completely resolved the problem. 

cjzimm76
Partner - Creator
Partner - Creator

So the upgrade to V11 SR2 did not have any effect on the missing data?

mongolu
Creator
Creator

Hello.

Had been your problem resolved?