Skip to main content
Announcements
Join us at Qlik Connect for 3 magical days of learning, networking,and inspiration! REGISTER TODAY and save!
cancel
Showing results for 
Search instead for 
Did you mean: 
aj96
Contributor III
Contributor III

Archive tables are not getting dropped and incorrect ETL instructions

I'm facing an issue where when I give drop and create in data warehouse Archive tables are not getting dropped and when I try to generate the code it shows that the object already exists. So I had to drop those tables manually and when i generate no error shows up but when you check the archive table instruction for copying ct record to archive, the select statement along with column names in the instruction is missing and I have to Clear the Metadata cache and generate again. This happens every time I drop and create all the tables in compose. 

 

archive table column missing 1.png

Labels (1)
7 Replies
Adam_Herman
Support
Support

hi @aj96 

Thank you for submitting your request.
The scenario which you have described requires a deep investigation of the issue.

Therefore, in case you're still facing this behavior, please raise a support case, and we will gladly assist you.

All the best,
Adam

shashi_holla
Support
Support

Hi @aj96 

Please let me know if you could try the option in Project settings ->  Reset Project -> Drop archive tables 

I think Drop and create in Data Warehouse might not delete archive tables since it's not exactly a DW object.

Thank you,

aj96
Contributor III
Contributor III
Author

@shashi_holla @I agree that droping dw doesn't need the archive table to dropped but error when generating should have been handled by the tool and having to do a workaround like reset is risky if someone click the wrong tickbox

@Adam_Herman Raising a ticket only wastes my time. Each new feature release is an headache because it's broken one way or another. I would rather not use these broken features than raise a ticket. These should have been raised in your testing and I'm not gonna set up environment and test for you guys to realize your mistake. Please make effort to test each release from your end.

shashi_holla
Support
Support

Hi @aj96 

I understand that since the Generate option is throwing the error on archive table, that the expectation is that Compose should drop the archive table also. But archive tables are part of the landing zone settings and Compose don't exactly interact with it similar to the other DW tables. Hence the additional option on the Project settings to remove the archive tables. If the archive tables were getting deleted as part of DW Drop and recreate then there was no need to provide this additional option.

Hope this makes sense and help you with your current issue.

Thank you,

aj96
Contributor III
Contributor III
Author

@shashi_holla No it doesn't make sense.

The archive tables are created when the cdc etl are generated. it should check if the tables are already there and ignore create and do the rest of generate. Even if the tables are dropped there is the invalid instruction issue.

Instead of giving me workaround for the issue which i already know, please accept this is a bug and fix it and don't bother telling me raise a support case becauseit never helps and a waste of my time.

You guys should try using compose and see how frustrating it is to use it daily when there are multiple deployment. 

shashi_holla
Support
Support

Hi @aj96 

Please let me know the Compose version you are on. The purpose of asking a support case is that we have a reference to go with and we can collect relevant details to help with the analysis. Right now we don't have visibility on what is happening on your side.

Thank you,

jtompkins
Creator
Creator

I agree there seems to be minimal regression testing and QA testing on Qlik's end. As an end user, I feel like a beta tester, only unpaid. It is extraordinarily frustrating. And when I open a case, often the first response is "did you try upgrading to the latest version?", when doing so results in even more problems. It's a constant struggle. Glad it's not just me.