Tracking the input file name when using the Generic Indexer

[Douglas]

Hello

I am using the Generic Indexer and JCL to call ARSLOAD via the USS (Unix System Services) directory in ZOS to load into CMOD 9.4.0.5. 

The indexing and loading is working fine.

My question is; the “System log” does not log the original name of the input file, is there a way to track the original input file name after it is loaded?

For example:

I am storing the following dataset in CMOD
 FILERST.YEAR05.F1DD.R160201

 After the load, the System log states: (same with System Load folder)

Processing file >/home/#1111/FILERST/FILERST
<Load Version <9.5.0.3>  Operating System <z/OS> <04.24.00>  OS Userid <1111>  Install Location </products/ars/val/> Data(unlimited KB) Stack(unlimited KB) Core(8192 512-blocks) Cpu(5402 seconds) File(unlimited 512-blocks) Nofiles(131072) Threads(0) Processes(0)
Server Version <9.5.0.1>  Operating System <z/OS> <04.24.00>  Database <DB2> <11.01.0000>
Application Group >AGFILERST<
Application >FILERST001-V01<
Storage Set >OAM01<
Storage Node >ONDMD.OAMVAL01.GOV01P01.MOVP001<
Loading started, --UNKNOWN-- bytes to process
OnDemand Load Id = >6755-7-0-7FAA-20050103000000-20051118000000-6756<
Loaded 100817 rows into the database
Document compression type used - OD77.  Bytes Stored = >39895820< Rows = >100817<
Loading completed
Processing successful for file >/home/#1111/FILERST/FILERST<

[Alessandro Perucchi]

Have you activated "System Load" ??
Because the "System Load" keeps track of that information, with a lot others.

[Douglas]

yes the "system load" contains the name of the index file,  not the input file. same as the system log.

[Greg Ira]

No way to track the original input file that I know of.  Best I can think of, if you have control over it, is to name your index file (and trigger file if using the daemon) the same as the input file.

[Alessandro Perucchi]

Hi Douglas,

Oh I see what you mean... no, for CMOD the important thing is the index name, never the filename. The files are just a container of the data, and all the information needed to find the "uniquely" file is in fact in the metadata or Indexes.
You know the filename is irrelevant for archiving, it's just the name of the container. What is important for an archive is the metadata. And don't forget CMOD is not a DMS, where you track the files, where you can version it, where you work at the file level... then in that case, yes it is important. But CMOD is not like that.

So, if the filename is important for you, then you MUST add a new field in the metadata in order to keep it.

Something like:

Code: [Select]

...
GROUP_FIELD_NAME:filename
GROUP_FIELD_VALUE:my_important_filename.extension
...
GROUP_FILENAME:my_important_filename.extension

And it means, you should also have this field in your AG.

Hope that helps to understand the philosophy of CMOD.

[Douglas]

Thank you all for your responses.