IBM Mainframe Forum

I have a step which gets the records

myddname dd disp=(old,delete),dsn=*.unldstep.smfunld.outdd.
how do i proceed towards getting that written to the existing gdg i have.
i want to be able to create a generation like xxxxxx.something(+1)
and then when i am in the utility panel, 3.4, it shows all the list of datasets starting with xxxxxx.

thanks

First of all, watch the DISP of your dataset. (OLD,DELETE) is "read once" only.

Are you saying you have a dataset which you want to copy to a GDG? You seem to have generally the right idea about it, so what is the problem you are having?

Hello and welcome to the forum,

Are you running some process to unload some smf data? Is the unloaded data for your use only? If so, why not create the new generation during the unload (i.e. the +1)?

If the unload file is not your own personal file, you need to remove the (old,delete) - whoever does own this may be highly displeased if you delete the dataset.

If you have not done so already, you need to define a new Generaton Data Group. Suggest you speak with your storage management people for the proper way to do this on your system.

This most likely is a task performed by an experienced systems programnmer, not by a newbie. It is generally done like this to retrieve SMF data from the SMF datasets.

The JCL manuals have examples of creating GDG datasets, and there is no difference for datasets that contain SMF data other than the somewhat unusual DCB attributes used by these datasets.

Most shops have prescribed rules about this; since we do not know these rules we probably should not assist you.

Thanks for the information.

Here is what I have



I dont know what comes next. I have my generation data group defined and exists , it is XYZS0000.ACCOUNTLOG.
There has be something like (+1) at the end that write it to a generation. This SMF dump data I meant.

OK, so from what I can tell, you have a temporary dataset "DSN=&&SMFTEMP" being created in this step. Following good coding practices, it would make sense to now add another step, checking that your step UNLDSTEP ended successfully, and then copying the temporary dataset to the GDG. What is the issue?

FYI, "XYZS0000.ACCOUNTLOG" can't possibly be a valid GDG base name.

It gives a JCL error because something is wrong with the syntax.
No it is the valid one.

The GDG base (XXXXXX.ACCOUNTLOG) is an entry in the catalog,
not a dataset in its own right, so it doesn't exist on any particular
volume: ISPF 3.4 indicates such entries with "??????". (It does the same
for other catalog-only constructs such as VSAM clusters etc).

"A generation data group (GDG) base is allocated in a catalog before
the generation data sets are cataloged. Each GDG is represented by a GDG
base entry. Use the access method services DEFINE command to allocate the
GDG base."

Each section of a dataset name cannot exceed 8 bytes - ACCOUNTLOG is 10 bytes - error

Ok for instance the GDG name is: