Hello,

I am trying to find out how to extract the files generated by Job Watcher in order to interpretate the information in an external tool. The problem is that I would like to automatize the extraction and the only way that I know is to download the files manually with the iDoctor interface. Do you know where the files are located?

Thanks,

------------------------------
Ignacio García Bieco
------------------------------

Dear Ignacio

I Google with "ibm i job watcher file name" and find Job Watcher's database file name in the URL on STRJW command:  Job Watcher collects data on a sampling basis and writes the collected data to a set of database files. The file names all begin with the letters 'QAPYJW'. 

To locate in which library these files are, you can use WRKOBJ command or use STRJW+F4 and look at the current value of its LIBRARY parameter.

You can download these files with File Transfer tool in ACS or SQL through an ODBC/JDBC connection, all of which you can automate. A complete list of QAPYJW files and sample extraction SQL can be found in this Job Watcher redbook:  https://redbooks.ibm.com/abstracts/sg246474.html   Keep in mind that the content of this redbook may not be up to date as it is now archived.

What kind of performance information you are looking for?  In my long experience with analyzing IBM i performance data, I find that using the standard IBM i performance collection services files (QAPM*) are useful for identifying the performance issue most of the time. I rarely needed to use JW data.  A list of the QAPM files is here :

https://www.ibm.com/docs/en/i/7.4?topic=csdf-collection-services-data-files-system-category-file-relationships
https://www.ibm.com/docs/en/i/7.4?topic=performance-collection-services-data-files

------------------------------
Right action is better than knowledge; but in order to do what is right, we must know what is right.
-- Charlemagne

Satid Singkorapoom
------------------------------

Hello,
What kind of performance information you are looking for?  As Satid already said, sometimes standard  performance collection services files are useful to identify performance problems or ti ave a wide view. 
If you're using iDoctor suite, and in particular Job Watcher collection, you have to consider that the JW collection store data within multiple files inside the library you've configured fot the collection. If you have multiple collections in the same library, each collection will have a member  inside the same files, so JW files are multi member files.
In general you can save and move your collection to another IBMi partition, upload to IBMfor an opened case or you can save the collection in other places into a SAVF. if you want to make a sort of automatic tool to save/export a JW collection, you can do the same commands that the iDoctor client do. The command is SAVJWCOL
You can find the command description at page 249 of the redbook linked by Satid, but you can also fint tons of documentation online.

If you perrform an analysis with the IDoctor client, you can also extract results as CSV file.




------------------------------
CADARIO SIMONE
------------------------------

Hello Ignacio, 

Remember you have at Pdi in new navigator the option to see Job Watcher data, then you can  click SHOW SQL and view what are the tables used ...