ViewZeroOut doesn't work

[EP_explorer]

Found one very strange thing in one TM1 model
I can't zero any view in one Cube in it using ViewZeroOut.

I've done different things - at the end I deleted all rules from cube and put in view only detail elements. But ViewZeroOut doesn't work again

CellPutN works, I can put 0 by hands but ViewZeroOut - doesn't

I even create test Cube in the model - ViewZeroOut works there. So it seems ViewZeroOut doesn't work only in the Cube which I need zero.

So any ideas - why ViewZeroOut can't work?

[ardi]

Found one very strange thing in one TM1 model
I can't zero any view in one Cube in it using ViewZeroOut.

I've done different things - at the end I deleted all rules from cube and put in view only detail elements. But ViewZeroOut doesn't work again

CellPutN works, I can put 0 by hands but ViewZeroOut - doesn't

I even create test Cube in the model - ViewZeroOut works there. So it seems ViewZeroOut doesn't work only in the Cube which I need zero.

So any ideas - why ViewZeroOut can't work?

Did you check if all subsets assigned to the view have the correct elements?

[EP_explorer]

I even at the end put detail elements in all dimensions. And result is the same. ViewZeroOut doesn't work.

Is is ppssible that cube in some way is corrupted?

[Steve Rowe]

Super unlikely....
Most likely you have a C level in your view, you can only zero out N level data and it must be explicitly in your view / sets.

Other reasons for confusion to arise.
It's ruled data and therefore can't be zeroed out.

Suggest you post your code and / or screen shot of view definition

[EP_explorer]

01.jpg (113.48 KiB) Viewed 19244 times

[burnstripe]

What version is this on
https://www.ibm.com/support/pages/flash ... hout-error

[EP_explorer]

What version is this on
https://www.ibm.com/support/pages/flash ... hout-error

It seems you are right
Version is 9.0.4
Thank you

[lotsaram]

What version is this on
https://www.ibm.com/support/pages/flash ... hout-error

It seems you are right
Version is 9.0.4
Thank you

Why are you using a 3 year old version?

[Alan Kirk]

What version is this on
https://www.ibm.com/support/pages/flash ... hout-error

It seems you are right
Version is 9.0.4
Thank you

Why are you using a 3 year old version?

3 wouldn't be so bad. Generally I don't regard sitting on IBM's bleeding edge releases to be... ah, shall we say, "prudent".

(He says, noting distinct performance degradation in ODBC (both in and out) between 2.0.6 and 2.0.9.18, and the fact that the transaction log in Perspectives often returns cr@p on the newer release, for example.)

But 9.0.4 is of course from 18 July 2008, making it 5 times that age. And to be fair there is a lot of good stuff (notably in the API) that has been added since then.

However that said... one does have to also consider the question of affordability, and a 15 year old TM1 is still better than no TM1 at all.

[EP_explorer]

Sorry
Planning Analytics 2.0.9.4 of course
))

[EP_explorer]

What version is this on
https://www.ibm.com/support/pages/flash ... hout-error

It seems you are right
Version is 9.0.4
Thank you

Why are you using a 3 year old version?

One client uses it
Actually (not in the case) but in general - I don't like migrate to new version because first of all you have to check how all components of the models would work in new version and it could take a lot of time (and bring some sort of surprises like I've found with ViewZeroOut - and I remember how I had to change command ProcessError to ProcessQuit in many processes in new version of PAL because suddenly ProcessError started to rollback all changes in cubes which proccess did )

[Alan Kirk]

It's not necessarily a case of "of course" above... I know that there are still a few V10 users around and I'd bet that the occasional V9.x user exists too. (I inherited a 9.5.2 instance only a couple of years back but I grant you 9.0.x would be way less common than that.) But that aside...

...first of all you have to check how all components of the models would work in new version and it could take a lot of time (and bring some sort of surprises

That's pretty much the point that I was making too, but of course we're both guilty of Old Skool Thinking. Now of course upgrades come out every few months and we should unhesitatingly trust that each upgrade is utterly flawless. Or, if you're on an IBM cloud environment, you don't have to trust at all; just be told how long you will be "permitted" to go without an upgrade.

After all... "What. Could. POSSIBLY. Go. Wrong?"

IBM has a STWATEGY!