Re: Long transactions
If you are on unix, use split command to split ur file to 1000 lines each and
then load
mosserp@wellsfargo.com wrote:
> Two other options:
> 1) set up an HPL job; you can run it from the command line (if you really
> don't want to use dbload, then this would be my recommendation).
> 2) switch your database to non-logging, run your load (leave out the
> "commit"), switch database back to logging, then immediately perform a
> level-0 archive.
>
> HTH,
> Paul Mosser
>
> -----Original Message-----
> From: Jay Aymond [mailto:jaymond@communitycoffee.com]
> Sent: Thursday, July 03, 2003 1:37 PM
> To: informix-list@iiug.org
> Subject: Re: Long transactions
>
> You could drop the indexes, alter the table to "raw", do the load, and then
> alter it back to "standard".
>
> "mpw" <em-2@cox.net> wrote in message
> news:lmXMa.564558$vU3.107079@news1.central.cox.net...
> > If I'm using a dbaccess script to load data from a file is there a way I
> can
> > tell the load to commit the transaction after every 'x' rows processed.
> I'm
> > trying to load a file with about 180000 records. I know I can use dbload,
> > but with the automated tools I have dbaccess works better.
> >
> > My script file is:
> >
> > database db1;
> > load from loadfile.dat insert into table1;> > commit;
> >
> > Thanks. M
> >
> >
>
> sending to informix-list
sending to informix-list