Re: truncate randomly failing

From: Matthias Rogel <rogel_at_web.de>
Date: Fri, 13 Dec 2002 16:29:02 +0100
Message-ID: <atcubv$12io17$1@ID-86071.news.dfncis.de>

Vladimir M. Zakharychev wrote:
> "Joel Garry" <joel-garry_at_home.com> wrote in message
>
>

>>Users view of sys.*$ is confused, because DDL is not transaction
>>based.

>
>
> Hmm... Are you saying that DDL violates ACID properties of the
> database? Or do I read this sentence incorrectly?
>

well, what is your opinion to the following:

as Jonathan Lewis pointed out, a truncate at first implicitly performs a commit.
After the commit it still may fail (in case another user has meanwhile performed a lock on the table)

in my opinion, the A in ACID is violated A = atomicity

Matthias Received on Fri Dec 13 2002 - 09:29:02 CST