TM1 Forum

Guys..

This is my first post in the TM1 forum.

I am running one TI process which we are using to generate some currency related flat files , every time this process suppose to generate 7-8 flat files and this process is scheduled in the chore, earlier it is used work and generate required no of files m but last few days its generated only 2 flat files and that too are not with full records after some particular record process terminated .

When i checked for TI process error log , it says Evaluation stack error for which i found one ibm technote which says there some problem with code but when i am running the same process manually it runs successfully with full records and all 7 flat files.

Ca anyone please tell me why this is happening when i am running same process manually it runs fine and same with scheduled chore it produce only 2 files with half records.

Hello,

but when i am running the same process manually it runs successfully with full records and all 7 flat files.

What do you mean by running the same process manually? What do you do when you run the process manually?

You can paste your code into this forum and into a post but make sure you do not sure sensitive information such as networks and paths.

If you paste your code in, people will be able to help you a lot easier.

Thanks,

Trevor

What i mean manually ,is run TI process by opening it , by clicking run button . unfortunately , i can not paste my code . what i can tell is there is one master process which executes bunch of process and this master process has scheduled in the chore .

So when this this chore runs with all these process , there is one process which generates flat files this process triggers and generate files but not with full data .in the contrary if I runs these processes manually they are executed successfully with full data export in the flat files.

can you tell me what i need to do .

What i mean manually ,is run TI process by opening it , by clicking run button . unfortunately , i can not paste my code . what i can tell is there is one master process which executes bunch of process and this master process has scheduled in the chore .

So when this this chore runs with all these process , there is one process which generates flat files this process triggers and generate files but not with full data .in the contrary if I runs these processes manually they are executed successfully with full data export in the flat files.

Ahh ok, so you are running a Master TI that calls separate TIs? You could use the error checking code in the IBM manual for Process Executing within another process http://www-01.ibm.com/support/knowledge ... uteprocess

You can create logs for each possible error for each individual process call.

Have you also taken security into account? Check the user who created the chore has the rights to fire them, but the link above gives good examples of how to check individual process calls.

There is no error if runs it manually . it only gives evaluation stack error when runs by chore.

and for evaluation stack i found below technote.

http://www-01.ibm.com/support/docview.w ... wg21457317

I want to ask you that same bunch of processes i am running through chore gives above error and generate files with half data but if i run the same TI processes manually they are running fine.

So can you suggest why they are producing error when run by chore and not by manual .

There is no error if runs it manually . it only gives evaluation stack error when runs by chore.

Is the Chore running each process separately? Or is the Chore running a Master TI which calls each process separately?

Thanks.

Trevor.

The chore is running master TI which runs each process separately.

The chore is running master TI which runs each process separately.

Is the timing of the Chore necessary? If not, try running the chore at a different time.

My thinking behind this is the possibility of running out of memory space.

sagar.ekhande wrote:The chore is running master TI which runs each process separately.

TI is no different from pretty much any other programming language. Each "function call" or ProcessExecute call will consume stack space of some amount.

If you are running just the single TI that generates the files manually, but the chore is running a different master TI you're not doing an apples-to-apples comparison.


If this is a new issue, what's changed recently? Chore definition? TI code?

You could also use Ops Console to monitor the chore whilst it is running and you can turn on Performance Monitor to view the memory consumption.

TrevorGoss wrote:You could also use Ops Console to monitor the chore whilst it is running and you can turn on Performance Monitor to view the memory consumption.

I don't think that's going to help anything. The problem is he's running out of stack space, not physical memory, and I'm pretty sure stack space isn't anything that you can monitor with any of the available tools out there.

I'm pretty sure stack space isn't anything that you can monitor with any of the available tools out there.

Indeed

I don't think that's going to help anything.

He could view the behaviour of his TM1 service as a whole to look for a potential cause.

Thanks guys ... i will surely try above all options.