Mainframe, MVS and zOS Discussion

Hi,

I have started working at a company. In this company, we keep on getting this email like, a given CICS region needs to be shutdown as a given file which is allocated to it is full.

It happens every now and then, actually nearly everyday for one region or other. Is this normal for a company which uses a lot of CICS? Or are there some better ways to handle it?

When I ask with my colleagues they said, we do it this way only.

Is this normal for a company which uses a lot of CICS?

No, most sites I've worked at (a) don't have any need to bring down regions for "full file", and (b) only send emails for critical issues (such as a region going short-on-storage).

Or are there some better ways to handle it?

Since yours is the first time, in more than 40 years of mainframe work, that I've heard of a need to bring down a CICS region for "full file", I'm sure there are better ways to handle the situation. Every site I've worked at has been able to keep CICS regions up for at least a week at a time (often, several weeks) with no loss of functionality.

Questions to ask:
- Is there a need for the CICS regions to be up 24/7 or can they be brought down at some point for the file reorganization?
- What, exactly, causes the "full file"?
- Is the file VSAM (and if so, is it ESDS, RRDS, KSDS)? If not, what kind of file is it and why does it get full (z/OS uses secondary extents when files run out of space, so the file should be able to extend itself unless it is direct access)?
- If the file is direct access (BDAM) or VSAM RRDS, why hasn't the application support group done an analysis to determine how many blocks to give it to prevent it from running out of space?
- For a given region, how long is the time interval between file reorganizations?
- How many regions?

I'm sure you can come up with other questions, but that should be a start.