Re: WDSCI-L Digest, Vol 8, Issue 357

[Jon Paris <jon.paris@xxxxxxxxxxxxxx>]

Yes to the first question and no to the second.

I had done some more testing based on Thierry's comments (Thanks  
Thierry) and it is certainly the variables view that causes it.

I suspect that this bug (in some flavor or other) may have been around  
for a while - it is just worse in this release. That would explains a  
number of weird issues of the debugger racing to completion (sometimes  
without stopping at all) we have had with debug but which we could not  
reproduce at will.

I did discover that once I was into an RPG IV program and had the  
variables view hidden, then step return etc. all worked just fine. I  
also tested bringing the view back and seeing what happened. It  
appears that everything works fine (variables updated etc.) as long as  
you stay within the bound PGM object. The minute I called to another  
program (in this case and RPG400 program) then the debugger just ran  
to the end ignoring all breakpoints etc.

So - until this is fixed the safest way seems to be to just close the  
variable view and pretend it doesn't exist!


Jon Paris

www.Partner400.com
www.SystemiDeveloper.com



On Nov 12, 2010, at 1:00 PM, wdsci-l-request@xxxxxxxxxxxx wrote:

> One problem (APAR number: SE44536) currently under investigation is  
> that
> the debugger hangs if variables view opens and
> you step back from a subprocedure to the caller.
>
> Did you set breakpoint inside a subprocedure ? Did you also hit this
> problem when the variables view was closed ?