Re: Impact of the year 2000 on Oracle systems

I don't believe there are year 2000 problems PROVIDED you do not rely on the Oracle default date mask. But here's a tip from the field!

I recently checked out a system of around 200 V3.0 forms under Version 7.1, all of which worked fine UNTIL someone entered a date of 29-FEB-00. (The rule is that years ending in 00 are NOT leap years unless they are also divisible by 400, ie 1600 AND 2000) Hence, 29-FEB-(20)00 is valid and was accepted at ON-VALIDATE level, BUT the forms as part of subsequent process passed 29-FEB-00 to a procedure as a DATE type. Oracle then applied its standard date format (ignoring the century) and collapsed as it attempted to validate 29-FEB-(19)00. Everything worked fine when the proacedure and calls were modified to pass as a CHAR type and then converted using YYYY. There may be ways of changing the default format ,,, I guess the message is to double check the 29th of February ... at least if you get it right, you won't have to worry for another 400 years!!!!

Bob Johnson <no-email_at_links.into.here> wrote in article <01bc1972$6030ac60$b92e64a1_at_azkrdw00173.krd.az.bp.com>...
> G'day All,
>
> A work colleague has asked me to post the following.....
>
> 'I am conducting a Year 2000 study for my client and am at the stage
 where
> I am looking at legacy applications and development tools.
>
> Could anybody give me any WebSites or NewsGroups that have recorded the
> experience with respect to Oracle development tools, applications and
> RDBMS's and/or if you have time could you reply with your experiences /
> opinions?'
>
> 'I had heard that the RDBMS is OK but the PL/SQL date routines are a bit
> suspect, is this true or not? What versions of the stuff are in doubt?'
>
> Replies to bj_at_iinet.net.com or johnsob1_at_bp.com or to this NewsGroup
>
> Thanks in advance to all who contribute.
>
> Bob Johnson
> Oracle DBA
> BP Refinery
> Kwinana; Western Australia
>
  Received on Thu Feb 13 1997 - 00:00:00 CET