RE: CHGJRN

["Andy Nolen-Parkhouse" <aparkhouse@xxxxxxxxx>]

Peter,

What you are describing is very important.  While it may be necessary to
find a work-around, I would encourage you to open a problem with IBM.
Journaling is the basis for system recovery in a number of scenarios and
you have described a gaping hole in its reliability.  If a transaction
cannot be recovered or recreated using a journal, what good is the
database?  Unless journaling is ended, _all_ changes _must_ be
journaled.  If IBM thought they were doing a good thing by having the
CHGJRN command process asynchronously, perhaps they need to rethink.

Regards,
Andy Nolen-Parkhouse

> On Behalf Of Peter Colpaert
> Subject: Re: CHGJRN
>
>
> Mark,
>
> I was under the impression that, as long as you don't submit a
command,
> the
> actions of this command are finished at the moment the next command is
> executed.  That's my idea of "immediately".
>
> The updates are not in the old receiver, in fact they are not found
> anywhere, but the file shows them, so they _have_ been done.
>
> Also which message should I monitor?  The online help only lists CPF
> *escape messages, which would crash my program anyway.
>
> Thanks,
>
> Peter Colpaert
> Application Developer