ABAP Program analysis with se30 how to avoid time out ?

[former_member246786]

Hi Gurus,

my problem is that I would like to analysis the performance issues of an ABAP Specific transaction.

BAsically, I launch se30, provide transaction code under transaction field and click on execute.

This launch transaction and then I put parameters and :

a ) launch transcation

b ) launch in background immediate.

In both cases it doesn't work due to time out and I never obtain a usefull result.

My question is : how can I use to analys such a transaction and avoid time out problem ?

Thanks for your help,

Regards

Morgan

[former_member246786]

Hi Gurus,

my problem is that I would like to analysis the performance issues of an ABAP Specific transaction.

BAsically, I launch se30, provide transaction code under transaction field and click on execute.

This launch transaction and then I put parameters and :

a ) launch transcation

b ) launch in background immediate.

In both cases it doesn't work due to time out and I never obtain a usefull result.

My question is : how can I use to analys such a transaction and avoid time out problem ?

Thanks for your help,

Regards

Morgan

[ThomasZloch]

One way would be to limit the selection criteria in such a way that the processed data is still representative but it finishes before the time out.

Even when a time out occurs, there should already be a trace file up to that point that you can evaluate. Depends on your processing logic, if it is one big outer loop it might be valid data, if you have several independent blocks, then you might be missing something important.

There is also the option "in parallel mode" where you can capture running BDC processes for analysis, maybe try this option as well. The trace file might grow pretty big though the longer your process runs, depending on aggregation level.

Thomas

[former_member246786]

HI Thomas,

thanks for your help.

I really think that I nead a complete analysis without parameter restriction.

I'm not sure to understand how to you parallel mode and how it helps me ?

Is there not a way to set some parameter or so to bypass the time execution for analysis ?

Is there other transaction that allow it ? st12 ?

Regards,

Morgan

[ThomasZloch]

"parallel mode" -> actually called "parallel session", please see the button on the right of SE30 entry screen. Start your program in background, go to SE30, hit the button, choose the running background process.

You can ask your system administrator to temporarily increase system parameter "rdisp/max_wprun_time" for your exercise. You might get turned down though.

Thomas

[former_member246786]

Thanks again,

what does "rdisp/max_wprun_time" parameter control ? Is it the max allowed execution time of a job in backgound or the max execution time of a foreground execution ?

Cause I thank this is foreground but the value in my system is 1800 and dump is before 30 minutes ...

Or maybe I do not understand well, this value is for background job but I just ask me to increase it before launching background job to avoid time limit while executing background job ?

[Former Member]

you should set the aggregation to 'calling position'.

You steup sounds looks like no aggregation which will always lead to an overflow with a batch.

[Former Member]

Hi,

try to Use tanscation SAT and check

D@s

[Former Member]

You may create your own measurement restrictions variant. Specify maximum runtime / max. size of file and execute with this variant.

[Former Member]

the SAT is based on the SE30 kernel function and behaves in the same way.

aggregation agggregation aggregation .... clear !!!

[Former Member]

You can Extend the time limited defaultly in your SE30. follow this .

SE30-> In 'Measurement restrictions' window->Click 'Create icon ->Go to 'Duration/type' tab-> there change the 'Maximum runtime'