Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Thanks I know the risk, this is only for my test database, not for production, don't worry. =0D =0D Juan Carlos Reyes Pacheco=0D OCP=0D -------Original Message-------=0D =0D From: Mark W. Farnham=0D Date: 10/06/04 13:08:28=0D To: jreyes@xxxxxxxxxxxxxxxx=0D Subject: RE: Please a parameter to disable undo, like _disable_logging=0D =0D too long a thread to read the whole thing.=0D =0D Did anyone point out to you that the difference is that a failure with=0D nologging or append will cause you to lose that individual object, while = if=0D your disable logging and space management takes places you may well lose = the=0D whole database?=0D =0D Since it is a whole lot easier to do and is supported I would recommend=0D either create as .... nologging or using /*+ append *+/ depending on what= =0D you are doing, rather than attempting to disable logging on the entire=0D database.=0D =0D Then if something in the load/create fails, you just clean up a bit and d= o=0D it again.=0D =0D After the fact you take a backup if you care about the results.=0D =0D mwf=0D =0D -----Original Message-----=0D From: oracle-l-bounce@xxxxxxxxxxxxx=0D [mailto:oracle-l-bounce@xxxxxxxxxxxxx]On Behalf Of Juan Carlos Reyes=0D Pacheco=0D Sent: Tuesday, October 05, 2004 8:23 PM=0D To: oracle-l@xxxxxxxxxxxxx=0D Subject: RE: Please a parameter to disable undo, like _disable_logging=0D =0D =0D thanks=0D =0D but /*+ append */ too bypasses undo, and if you set _disable_logging you = run=0D the same risk, if you have a problem.=0D I think for a test database could be acceptable. For example I'm trying t= o=0D create a 10 000 000 records in my home for testing=0D purposes, so this will be greatly welcome.=0D I bet there should be a way.=0D =0D =0D Juan Carlos Reyes Pacheco=0D OCP=0D --=0D //www.freelists.org/webpage/oracle-l=0D =0D =20 -- //www.freelists.org/webpage/oracle-l