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

Conference orarep::nomahs::rdb_business

Title:rdb_business
Moderator:NOVA::FEENAN
Created:Mon Jan 09 1995
Last Modified:Thu May 29 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:153
Total number of notes:620

147.0. "Do we test dynamic lock remastering ?" by NOMAHS::SECRIST (Rdb WWS; [email protected]) Mon Feb 17 1997 08:53

    
    	Does any of the Rdb product family's regression testing include
    	scenarios involving dynamic lock remastering ? 
    
    	Regards,
    	rcs
    
T.RTitleUserPersonal
Name
DateLines
147.1I can not think of any casesHOTRDB::LASTOVICAIs it possible to be totally partial?Mon Feb 17 1997 09:272
    Not that I'm aware of.  We tend to test Rdb and let Digital test
    OpenVMS.
147.2HOTRDB::PMEADPaul, [email protected], 719-577-8032Mon Feb 17 1997 11:442
    Well, not intentionally, but we do sometimes test in clusters, which
    implies that lock remastering is likely to occur.
147.3Those wonderful defaults...NOMAHS::SECRISTRdb WWS; [email protected]Mon Feb 17 1997 12:2512
    
    There is nothing on record that says we do or do not support
    dynamic lock remastering though, is there ?  It is hard to
    believe that anyone would want to move huge lock trees around
    in the middle of the day on a production system on purpose
    (but that is the VMS default) -- let alone the ramifications 
    of that in terms of software complexity.  In the back of my
    mind I always hear Dr. McCoy ranting about the transporter ;-)
    
    Regards,
    rcs
    
147.4HOTRDB::PMEADPaul, [email protected], 719-577-8032Mon Feb 17 1997 13:264
    Why should we document whether we do or don't support a particular VMS
    internal mechanism?  If the mechanism doesn't work then VMS engineering
    should fix it.  It isn't up to us to second-guess the usability of our
    platform operating system services.
147.5delete meNOMAHS::SECRISTRdb WWS; [email protected]Tue Feb 18 1997 11:223
    
    If we produce a repeatable case which demonstrates a problem in
    VMS functionality that adversely affects Rdb