HP3000-L Archives

August 2001, 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:
Michael Berkowitz <[log in to unmask]>
Reply To:
Michael Berkowitz <[log in to unmask]>
Date:
Fri, 3 Aug 2001 08:28:32 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (48 lines)
Ronald Horner writes:

Sure it can.  You abort a job at the wrong time, and you may be looking at a
damaged database.

Gibson Nichols wrote:

> A programmer asked me if doing an ABORTJOB on a program will cause a
broken
> chain in an Image database.
>
> Also we've seen cases where one record out of two will be on a database.
We
> suspect that the second record didn't make it because of an ABORTJOB.  In
> this case I was attempting to tell the programmer about the concept of
> logical transactions and using DBBEGIN and DBEND.  The programs are not
> written with the concepts of logical transactions.  They just write one
> record then a second.
>
> So, does using ABORTJOB cause IMAGE problems?
------------------------------------------------------
The answer is NO.  An abortjob to an application is no different than
pressing break and aborting a session.  In other words it is a clean
shutdown of a process.  This means files are closed, cached data is posted,
locks are respected, and all other aspects of a processed/job-session are
terminated in a normal fashion.  That being said, an abortjob can cause a
logical inconsistency in data.

Example:
DBLOCK to data set #1
DBPUT to data set #1
                    < :ABORTJOB done here
DBUNLOCK
                    < Abortjob happens here
DBLOCK to data set #2
DBPUT to data set #2
DBUNLOCK

In this scenario, abortjob will wait for DBUNLOCK to happen and then kill
the job.  However you are logically inconsisten because the second data set
did not get updated.  However there will be no broken chains.

Mike Berkowitz
Guess? Inc.

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2