Lester,

thanks for your answer.
I have never used the BDE and my project _is_ based on
IBObjects. It ran fine for about half a year (24 x 7)
until these errors occured.
There are no complex queries, no bad plans, no gaps
between OAT an OIT, no problematic sweeps, no corruption.

And when the IBSERVER.EXE is at 100% CPU usage for a long
time because of a client select, there must be at least
one client which is waiting a long time for the result of
the select.
But all the selects of my clients are completed in less than
a second. And the users don't have the opportunity to run
individual selects.

When hearing how many people ran into this problem, I am
beginning to believe that this is a bug.
I am really at the end of my rope now and the only solution
for me would be to put a second processor into my server and
to tie IB to this processor - so that the rest of the server's
services is not slowed down by this bug.

This is not very satisfying but what can I do?
At least I'd like to see that all people who ever had this problem
confirm it in this list after reading my message.
So it will perhaps be logged as a bug or make in the IB developer
handbook.

So please raise your voices!



Christian

> -----Ursprüngliche Nachricht-----
> Von: Lester Caine [mailto:lester@...]
> Gesendet: Dienstag, 12. Juni 2001 06:46
> An: ib-support@yahoogroups.com
> Betreff: Re: [ib-support] IBSERVER.EXE at 100% CPU usage for
> a long time
> - a bug
>
>
> I am another 'poor windows user' who has had to cope with
> this sort of problem. I think I have seen the same sort of
> effect in the past, and we are only using small databases (
> sub 10Mb ).
>
> In the last 6 months or so I don't think we have had the
> problem. That was since we ditched BDE completely - it is
> not even installed - and only run with IBObjects.
>
> Also we are running IB6, but both Borland and Firebird
> flavours - once it started running 24/7 we left it alone in
> the field, but at present Firebird seems to be quite stable
> on the soak machines, and the intention is to standardise on
> that.
>
> I never investigated the 100% utilisation because it would
> never happen when looking for it, but changes to the
> programs seem to have cured it because the site operating
> system and Interbase setup has not been changed. IBO
> stabilised a number of things that previously were a problem
> ( 100Mb growth in 15 mins on a 5Mb database ! )