[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference ulysse::rdb_vms_competition

Title:DEC Rdb against the World
Moderator:HERON::GODFRIND
Created:Fri Jun 12 1987
Last Modified:Thu Feb 23 1995
Last Successful Update:Fri Jun 06 1997
Number of topics:1348
Total number of notes:5438

109.0. "INGRES and v5" by BRILLO::BIRCH (Peter Birch, DTN 842-3297) Thu Mar 31 1988 17:26

    Hi,
    
    My customer needs reassuring that version 5 VMS won't do anything
    disastrous to his INGRES applications. He has a cluster of 1x8200
    and 2x8700 which he intends shortly to upgrade to 8800 (sorry 8820-N
    systems). Do we know if RTI have tested INGRES against v5 yet, and
    if so what the outcome was?
    
    Cheers
    
    PDB
T.RTitleUserPersonal
Name
DateLines
109.1Did they ask the wrong vendor?NOVA::CAMERONFri Apr 01 1988 20:146
    This sounds like a question for their Ingres support people.
    By the way, if they are a FT site for a future release of VMS, are
    they allowed to tell people how well they work with it?
    
    ~/~
    
109.2Yes but...BRILLO::BIRCHPeter Birch, DTN 842-3297Tue Apr 05 1988 16:1812
    To some extent I agree. However, it's not INGRES which is changing,
    but VMS, and therefore the customer sees it as DEC breaking RTI's
    product, if indeed it does. He basically needs a warm blanket; I
    can tell him that all user mode code works without change, performance
    hits are neglible, and so on, but he still asks 'what does that
    mean for ME?' Hence the question.
    
    PDB
                              
     
                                
    
109.3Ask RTIBANZAI::BERENSONRdb/VMS - Number ONE on VAXTue Apr 05 1988 16:4719
Digital really can not know if VMS V5 breaks INGRES or not, since we
don't know enough about INGRES.  If INGRES ran properly in an ASMP
environment (V3.x or V4.x of VMS on 782, 8350, 8800), contains no kernel
mode code (it probably doesn't), and contains no references to the VMS
Exec (which it might) then it should work on VMS V5.0 in an SMP
environment.  Otherwise, RTI will have to issue a new release to support
V5.0 and SMP (as do a fair number of Digital layered products).

Only RTI can tell you if they have already tested INGRES with V5.0 (for
example, if one of the VMS V5.0 field test sites are INGRES users).
Otherwise, RTI will have to wait until V5.0 is available, and then make
sure it works on V5.0.

There are a number of things, such as kernel mode code, that make
software non-portable between versions of VMS.  If a software vendor
uses these features, then they lag behind the availability of new
versions of the operating system.  One of the benefits of an all-Digital
software solution is that the lag time is either non-existant or else
very very short (measured in weeks).
109.4No, it does not run on VMS V550446::APPSRDB-S-IGN, Ignore possible bugcheck messagesWed Apr 06 1988 16:1010
The current version of INGRES (don't know the number) does NOT work on VMS V5.

I happen to know because an attempt was made to install INGRES on a V5 system
here in Germany.  The installation 'broke' when one of the INGRES routines was
to be "$INSTALLed" with privs and protected.
(Sounds as if they have a small piece of privileged code running above user
mode.)

John