Though I used share in both CL, I had deleted that override in that program itself after closing the file and then called the second program. I have commetned out the call to SQLRPGLE program, but still the error is there. But one thing I noticed was my Rcvf was not looping till EOF as for testing purpose I had commented out the loop to read the file again. So the RCVF was reading the file only once and then it was calling the RPG program, then close the file delete the override and call the second RPG program. Now I remove the comment and the file is reading till CPF0864 and the problem is resolved. But if the file is read only once….then even after closing the file why it is not allowing to do another opnqryf in the second program remains unanswered..

Correction… What is the SFLCTL DDS for fields (rather than the “subfile” itself)? — Tom

What computer? What is the OS? Is this your personal home system or a business system? How have you maintained old restore points? — Tom

The RPG possibly doesn’t matter (though the OVRDBF could make odd trouble; I’ve never mixed OPNQRYF and embedded SQL which seems almost contradictory).
.
The ODP is *SHAREd in the outer CL and also *SHAREd in the inner CL while the outer is still active.The outer ODP still exists. The two CLs have different OPNQRYFs, so the *SHARE probably should not be used at all. The two CLs need different ODPs. There’s nothing that should be shared as far as I can tell.
.
Tom

In 2001, the coding was already old and out of date. But today it needs to be converted soon. Tools such as SEU have already been stabilized and will not support new language features that have been added since i 6.1. And the old compilers have become a separate extra charge in i 6.1.
.
To provide your client best service value, you should make sure that your client knows they may be headed for some unexpected charges for upcoming system upgrades.
.
That’s separate from simply losing programmers who were familiar with the language due to retirements. And that includes IBM developers if support is needed. (Not to mention people like us trying to remember how things used to work.)
.
But again, is there a reason you don’t simply use the automatic support? You shouldn’t need to use the IN and OUT operation codes if you’re getting the original data area values and updating it. The program can get the data area automatically when it starts and and write the data area when it ends. Any changes your program makes will be stored automatically if that’s what you want. It might not be necessary to use *NAMVAR DEFN at all.
.
Tom

Please clarify. What do you mean by “windows”. The Mac OS uses windows in its presentation modes. Or are you asking about installing a Microsoft Windows OS on Mac hardware? And what “mac” are you asking about? What is the hardware? What OS is currently installed? And what do you mean by “best”? Do you want the cheapest? The fastest? The easiest to install? The easiest to use? (Easiest for whom? A general user? A developer?) The one that runs some Windows program that you want to run? (What program?)
.
Tom

Can we see the DDS that describes the subfile record fields? Their attributes might affect the useful ways to progress from field to field. For example, numeric entry fields or dates might be defined with attributes that interfere with automatic movement, but we can’t tell without seeing it.
.
Tom

The way I understand it, it’s not all that different from cloud computing. It’s merely defining, designing, implementing, and managing networks from the perspective of software rather than hardware. A physical decoupling so to speak.

Great discussion point Chris…I see some people *try* to backup everything. That’s an exercise in futility and can create unnecessary expense and liability. I also see people assume they’ve backed up everything that counts but they’re not even close.

The problem here is complexity associated with unstructured information. People simply don’t know what’s where. If they believe they do and it’s pointed out that critical information is also being stored somewhere else, the response is always the same: “Oh yeah, I forgot about that.” I don’t envy network managers these days.

More info on finding and securing unstructured information here: http://www.principlelogic.com/storage.html

In one AS400 shop I was in, the only thing that was “critical” that was lost, was my own development library.
Ever since then (over 20 years ago), I have never bothered to define what is critical. My backup stratagey uses logic that backups up everything that I have determined is NOT critical.
Thus when some new library or folder, etc is created, it will be backup up by default, unless I identify it as NOT critical.

i think Rajesh wrote that himself..lol 

Are you useing CHECK(FE)?
If so, that is forcing use of the Field Exit key.

Oh ok thank you.

If you write your own code, the most common way is to have the name of the output queue as input to the program. That was you do not do a global delete. There are legimate reasons to keep some spool files longer than 14 days.

The quick answer is NO

Many companies have user writen code to do this.
Another option is to use the expiration date on printer files. That would help you in the future but not for current spool files.

I’m still of the opinion that you open the queries at one level without a DCLF and work with the results in a called program.

But as far as deterimining if the RPG program is hanging up the close, just don’t call it an see if the error goes away.
Phil

Hi Tom!
you have very rightly said as to why to use an old programming(viz., RPG III), I can understand the frustration with it. Given a choice, even I would like to go for RPGILE.
but Iam helpless, as Iam changing an old program written way back in 2001, to add some more fields & validations to it. The client will not accept for the change in the language altogether. I don’t have a choice.
However thanks a lot guys!, for your patience & prompt replies.Really appreciate them. Thank you!

Tom,

The entire scenario is that I have a report that is displays info from a query. The report is called by a macro. If I run the query directly, no problem. If I open the report, the included parameter is requested but then one or more other parameters are asked for that are not part of the query. Weird or what?

wem3rd