HP3000-L Archives

August 1999, Week 3

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:
Tony Summers <[log in to unmask]>
Reply To:
Tony Summers <[log in to unmask]>
Date:
Thu, 19 Aug 1999 15:09:49 +0100
Content-Type:
text/plain
Parts/Attachments:
text/plain (18 lines)
>> A question has come up that we're sure someone on this list can help us
answer.  What happens if a user with a NS connection "kills" his terminal
emulator in the middle of a multi-record update?  With a serial connection,
this has not been a problem...the session stays alive.  But with a NS
connection, is the session and it's processes immediately terminated
potentially leaving some of the updates undone?  If so, is there a way to
prevent this?

>> reply >> 

Yes - this happens here when a user can't wait for the transaction to complete - so we try to ensure that all of our users have confirm on exit set in their terminal emulator before they then close it down.  Even this, however, can't prevent the longer transactions from being aborted mid-update. 

The only solution is to ensure your database/files/applications have sufficient Rollback/commit options - you need them anyway should an application or the system abort.  Our database is Ksam so we've built our own Rollback subroutines - I'd be interested to hear what happens in Image. 



==============================================================

ATOM RSS1 RSS2