HP3000-L Archives

February 1995, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Elbert E Silbaugh <[log in to unmask]>
Reply To:
Date:
Fri, 3 Feb 1995 15:38:55 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (33 lines)
Leslie-Anne writes about SQLMON and wanting addiitonal info...
 
I don't yet have the stuff to play with, but I have learned that to
find a solution to a problem you need
to be able to have enough trace/status information about the problem
so you can analyze and then abort, correct, signoff, reboot or
(final resort) kill the bloody person.
Sometimes you start with hunches. Sometimes you need to start the
resolution process with no assumptions.  i.e. I Do not know for sure
who is doing ODBC (or whatever) from XYZ pc in Timbuktoo bldg.
It means that we need to be able to not just abort the offending
process, but to determine specifically who the guilty party was/is
and to find out what they were doing.
With that info, you then also attempt to make changes so it does not
happen again.
 
I would expect that SQLMON and/or other tools will/should be able to
provide this information.  If SQLMON cant provide adequate info in the
lab, then it will definately not provide enough info in the field.
 
As far as novice or super user is concerned, from the PC viewpoint, I
find that both are capable of asking for queries that dont make sense
or end up asking for a zillion records. Try a join query on 2 tables
with 30000 records each -- but you forgot the join line! (You expected
100 records.) This is one reason some of us have always told the big
cheezes up above why we cant afford to let a user loose with a 4/5th gl
product with no control. Alas, the train has come and continues to
roll on....
 
I hope this general response helps.....
 
Elbert

ATOM RSS1 RSS2