Alex,

<< How long have you been aware of this.  In conversation previously when I
was asked why we were still on 3.21, I said because it was stable, and no
mention was made that Index Page Buffers could be an issue. >>

It was found in version 4.x long after 3.x was frozen.  Also, what previous
conversation are you referring to ?

<< This seems fairly serious especially since a rebuild seems to delete
actual data on ccasion.  Anyway we are desperate for this new dbisamem.pas
unit to fix this critical issue.  Our prgrammer is going on leave soon and
we need to resolve. >>

It was sent to you earlier this afternoon (EST).

<< Also are there any other potential issues you are aware of in a high
volume environment? >>

No.  But you also have to realize that we haven't touched 3.x in at least a
year, and we don't go back and re-test 4.x bugs against 3.x to see if it
would be affected.  This is why we are so adamant about people upgrading to
the latest major version in a reasonable time frame.  We simply don't have
the resources to do active development on multiple major versions for long
periods of time.  3.x was even longer than normal in that we issued 3.x
updates for quite a while after 4.x was released.

--
Tim Young
Elevate Software
www.elevatesoft.com

Tim,

<<
No.  But you also have to realize that we haven't touched 3.x in at least a
year, and we don't go back and re-test 4.x bugs against 3.x to see if it
would be affected.  This is why we are so adamant about people upgrading to
the latest major version in a reasonable time frame.  We simply don't have
the resources to do active development on multiple major versions for long
periods of time.  3.x was even longer than normal in that we issued 3.x
updates for quite a while after 4.x was released.
>>

Could you also send this to me?

We have a medium sized application running in v3.30 that we were going to move to v4.x, but then heard that v5.x was on the way. We have the
same issue as you in terms of resources. It would take quite an effort to QA all the SQL scripts (a couple of months) to make sure they work
properly under v4.x.

Can you give some guidance on the v5.x schedule? If it is still 3 to 6 months off, then we probably need to move our app to v4.x as we just
couldn't afford a corruption issue like the one describe in this thread out in the field.

On the other hand, if v5.x is targeted for release (or preview so we could start porting) in the next 1 - 2 months, then this would really help out a
shop like ours where we are tight on resources

I'd appreciate whatever information you are willing to share.

Either way, I'd like the v3.30 source file that you provided to others on this thread.

Thanks,

-Kevin

Kevin,

<< Could you also send this to me? >>

Sure.  Just send me an email (timyoung@elevatesoft.com) and I'll reply with
the source file.

<< We have a medium sized application running in v3.30 that we were going to
move to v4.x, but then heard that v5.x was on the way. We have the same
issue as you in terms of resources. It would take quite an effort to QA all
the SQL scripts (a couple of months) to make sure they work properly under
v4.x. >>

The only difference of any significance between 3.x and 4.x in terms of SQL
INSERT/UPDATE/DELETE/SELECT is the single-quote/double-quote issue with
constants and the change of the MEMORY keyword to the "Memory" database
name.  Any other changes were fixes that 3.x was performing incorrectly, or
where minor changes to the CREATE TABLE/ALTER TABLE table
creation/alteration.

<< Can you give some guidance on the v5.x schedule? If it is still 3 to 6
months off, then we probably need to move our app to v4.x as we just
couldn't afford a corruption issue like the one describe in this thread out
in the field. >>

It's not 6 months off, but it is still at least a month or so off.  February
and March so far have not been very productive in terms of development time.
Unfortunately that's simply one of the bad sides of having limited
development resources.

--
Tim Young
Elevate Software
www.elevatesoft.com

Hi Tim,

Believe me I realise what limited resources means.  Like the gentleman Kevin
said, it takes a great effort to move from V3 to V4 even if it is just as
you say, quotes, as we have many SQL statements and reports that would need
to be changed and fully tested to ensure nothing is broken.  I have also
been waiting for Version 5.

My previous conversation was only recent (we moved to 3.3 from 3.21 only a
couple of weeks ago, and no mention was made of this serious bug, I am aware
that it could slip the mind and Joe couldnt find anything describing that
particular issue.  So I was just asking.

Thanks

Alex

"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message
news:5282549E-C7C2-4AC6-9493-9CB6563B4F77@news.elevatesoft.com...
> Alex,
>
> << How long have you been aware of this.  In conversation previously when
> I was asked why we were still on 3.21, I said because it was stable, and
> no mention was made that Index Page Buffers could be an issue. >>
>
> It was found in version 4.x long after 3.x was frozen.  Also, what
> previous conversation are you referring to ?
>
> << This seems fairly serious especially since a rebuild seems to delete
> actual data on ccasion.  Anyway we are desperate for this new dbisamem.pas
> unit to fix this critical issue.  Our prgrammer is going on leave soon and
> we need to resolve. >>
>
> It was sent to you earlier this afternoon (EST).
>
> << Also are there any other potential issues you are aware of in a high
> volume environment? >>
>
> No.  But you also have to realize that we haven't touched 3.x in at least
> a year, and we don't go back and re-test 4.x bugs against 3.x to see if it
> would be affected.  This is why we are so adamant about people upgrading
> to the latest major version in a reasonable time frame.  We simply don't
> have the resources to do active development on multiple major versions for
> long periods of time.  3.x was even longer than normal in that we issued
> 3.x updates for quite a while after 4.x was released.
>
> --
> Tim Young
> Elevate Software
> www.elevatesoft.com
>
>

Alex,

<< Believe me I realise what limited resources means.  Like the gentleman
Kevin said, it takes a great effort to move from V3 to V4 even if it is just
as you say, quotes, as we have many SQL statements and reports that would
need to be changed and fully tested to ensure nothing is broken.  I have
also been waiting for Version 5. >>

I understand.  However, the changes going forward are always towards greater
compatibility with ANSI standard SQL, not less, so you can take comfort in
the fact that we will be standards-compliant with version 5 in terms of any
SQL that we support.  That will ensure that future changes to the SQL will
only change if the standards change, which they rarely do in the key areas.

<< My previous conversation was only recent (we moved to 3.3 from 3.21 only
a couple of weeks ago, and no mention was made of this serious bug, I am
aware that it could slip the mind and Joe couldnt find anything describing
that particular issue.  So I was just asking. >>

My apologies for not catching your comment about moving from 3.30.  The
issue was in 3.x for some time apparently and I haven't gone back to compare
the sources to find out exactly when, so I sometimes make the mistake of
assuming that if you're using 3.x, then you probably already encountered the
problem if it was a problem.  It's actually quite rare, actually.

--
Tim Young
Elevate Software
www.elevatesoft.com