Transformer creating junk category when model is saved (Urgent)

[cognos05]

Hi All,

I have a model built and it has time dimension, the datasource has values from jul 2015 to jun 2016 and its a flat file.
Now the problem is when I built the model it all looks fine ,and the source value for Jun 2016 shows 062016, but when I save and reopen the model, the source value for Jun 2016 changes to some junk source value as 99991232.so Next time when categories are generated another jun 2016 will be generated and when saved , this will not reopen again.
I created a new model and that doesnt creates this junk source value 99991232. I tried deleting this junk from the model file, but it still creates every time when I save the model.
I have security set up in this model, so I cannot create a new model and this cube is in production.
I have a ticket with IBM , but no response till now. Any advise is appreciated.

I tried deleting the time dimension and recreated, Checked the source value and no duplicated for jun 2016 and works fine with a new model.
I think the file is corrupted and not sure how to recover it.

Thanks,
Nithya

[MFGF]

Hi All,

I have a model built and it has time dimension, the datasource has values from jul 2015 to jun 2016 and its a flat file.
Now the problem is when I built the model it all looks fine ,and the source value for Jun 2016 shows 062016, but when I save and reopen the model, the source value for Jun 2016 changes to some junk source value as 99991232.so Next time when categories are generated another jun 2016 will be generated and when saved , this will not reopen again.
I created a new model and that doesnt creates this junk source value 99991232. I tried deleting this junk from the model file, but it still creates every time when I save the model.
I have security set up in this model, so I cannot create a new model and this cube is in production.
I have a ticket with IBM , but no response till now. Any advise is appreciated.

I tried deleting the time dimension and recreated, Checked the source value and no duplicated for jun 2016 and works fine with a new model.
I think the file is corrupted and not sure how to recover it.

Thanks,
Nithya

A good way of fixing a corrupted binary model is often to save it as .mdl. Once done, open the .mdl and save as a new .pyj. Might be worth a try?

MF.

[cognos05]

There was a bug in transformer , the source value and latest date in time category had jun 2016, so transformer created junk category, the solution was to change the latest date to something like 2050   or upgrade to fixpack 3
the model was already saved in .mdl file.
Thanks MFGF for your help .
Thanks,
Nithya