"Parker, Kathy" <[log in to unmask]> wrote:
>We are running C.05.08 version of Image, HP claims we should be running
>version C.06.07 to handle the jumbo datasets. I would agree with this,
>but, per what I thought the definition of jumbo sets was (4GB), we don't
>have such an animal (ours was 2.8GB).
 
Kathy, I am sorry for the confusing information given to you regarding
TurboIMAGE version C.05.08. The Response Center should have explained that
C.05.08 is a rather old version, and if a NEW problem is found with this
version, and the problem is resolved in a later version, you would be asked
to upgrade to the later version. No sense in patching a version, when the
fix is already in a later version.
 
There has been numerous fixes and enhancements to the C.05.xx IMAGE class.
If you want to stay on the C.05.xx versions, then I would suggest you go up
to C.05.11 (the latest and last of the C.05.xx series). Otherwise, all new
fixes are going into the C.06.xx versions, of which C.06.08 (referenced by
patch TIXHXK1) is the lastest GRed version.
 
As far as your corruption goes, I hope the RC asked if there had been any
hardware problems, powerfails, etc, occuring at your site. Also, was it
determined that the broken chain messages were "real" vs. the not locking-
around-reads scenario? I have never heard of any broken chain problems being
caused by Superdex or ODBC.
 
Best regards,
Steve Patrick
HP Database Expert Center