HP3000-L Archives

February 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:
THOMAS COOK <[log in to unmask]>
Reply To:
THOMAS COOK <[log in to unmask]>
Date:
Wed, 17 Feb 1999 17:52:10 -0300
Content-Type:
text/plain
Parts/Attachments:
text/plain (39 lines)
As a SIGIMAGE member and DBA I will support (within reason) anything that
can resolve the "stalled transaction" situation which has occassionally bit
my behind.

I prefer OPTION #1.

Thank you very much

Thomas G. Cook

B T Vikram Kumar wrote in message ...
>Hello All,
>
>Currently, CSY database lab is considering an enhancement to increase
>the transaction size for dynamic transactions (ie, transactions with a
>structure like DBXBEGIN -
>DBPUT/ DBDELETE/ DBUPDATE - DBXEND/ DBXUNDO). And also to give some sort
>of user control over the transaction. One feature, which is being
>planned to implement is that, to issue a warning to user when
>transaction size reaches a pre-defined soft limit. Once this warning is
>issued, users can have two options. Option #1: User is forced to commit
>or rollback (DBXEND/ DBXUNDO). Option #2: Don't force anything, user may
>
>chose to rollback/ commit or ignore the message. If it is ignored,
>ultimately the transaction size may exceed the maximum possible value
>and get a stalled transaction (i.e. compulsory rollback and a process
>abort). Which of these options, you feel will be appropriate? Your
>valuable feedback on this will help the lab in deciding on an
>implementation strategy and coming up with the solution.
>
>You may choose to post the reply in 3000-L or mail to me at
>[log in to unmask]
>
>Thanks in advance
>
>Regards,
>Vikram
>

ATOM RSS1 RSS2