Re: Problem with Sequences in Oracle 8.0.5

Yes, I just eliminated the CACHE 20
Use is not very intensive, so I'll try the NOCACHE option - otherwise I'll try not to worry!

Thank you!

"Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> skrev i melding news:997098611.16455.1.nnrp-01.9e984b29_at_news.demon.co.uk...
>
> When you recreated it without cache' do you
> mean you recreated it using the term NOCACHE
> explicitly, or did you just eliminate the 'CACHE 20'.
>
> In the absence of a specific directive, the default
> is CACHE 20 - so my guess is that your sequence
> is still cached.
>
> If this is a high usage sequence, you don't really
> want to make it NOCACHE - the overhead gets
> a bit high. Also, even with NOCACHE it is still
> possible to 'lose' values in the data set (e.g. a
> user acquires the sequence, then rolls back
> the insert).
>
> Don't worry too much about losing values - it's the
> way sequences work: you can reduce the impact of
> the problem in your version of Oracle (I think)
> by adjusting an init.ora parameter called
> something like 'sequence_cache_entries' (which
> also happens to default to 20). I think 8.0.5
> also allows you to use the dbms_pool package
> to KEEP sequences, again reducing the problem.
>
>
>
> --
> Jonathan Lewis
>
> Seminars on getting the best out of Oracle
> Last few places available for Sept 10th/11th
> See http://www.jlcomp.demon.co.uk/seminar.html
>
>
>
>
> Eyvind Elvestrand wrote in message <9klvke$61b$1_at_snipp.uninett.no>...
> >Have an inventory database running on NT4/Oracle 8.0.5.0.0.
> >The GUI is programmed in Delphi 5 / Odac (ODI)
> >For continuous number series I use Oracles's sequence "machinery", like
> >this:
> >select SEQ_INV_NR.nextval from DUAL
> >
> >Now the select statement has begun returning higher numbers than
 expected -
> >typically 20 higher!
> >
> >Originally the sequence was created like this :
> >CREATE SEQUENCE SEQ_INV_NR
> >START WITH 214575
> >INCREMENT BY 1
> >NOMAXVALUE
> >NOCYCLE
> >CACHE 20
> >
> >I suspected the cache part, dropped the sequence and recreated it without
> >cache - no good.
> >
> >Has anybody got a clue!
> >
> >Best regards from Eyvind.
> >
> >
> >
> >
> >
> >
> >
>
>
Received on Mon Aug 06 2001 - 07:17:58 CDT