Path: dp-news.maxwell.syr.edu!spool.maxwell.syr.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!postnews.google.com!i39g2000cwa.googlegroups.com!not-for-mail From: thomas.kyte@oracle.com Newsgroups: comp.databases.oracle.server Subject: Re: truncate /*+append*/ Date: 3 Mar 2006 08:14:19 -0800 Organization: http://groups.google.com Lines: 45 Message-ID: <1141402459.659701.130480@i39g2000cwa.googlegroups.com> References: <1141328588.206530.61630@p10g2000cwp.googlegroups.com> <1141331016.239208.326430@i40g2000cwc.googlegroups.com> NNTP-Posting-Host: 148.87.143.4 Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" X-Trace: posting.google.com 1141402465 23570 127.0.0.1 (3 Mar 2006 16:14:25 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Fri, 3 Mar 2006 16:14:25 +0000 (UTC) User-Agent: G2/0.2 X-HTTP-UserAgent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.1) Gecko/20060111 Firefox/1.5.0.1,gzip(gfe),gzip(gfe) X-HTTP-Via: 1.1 inet-nc05 (NetCache NetApp/5.6.2R1) Complaints-To: groups-abuse@google.com Injection-Info: i39g2000cwa.googlegroups.com; posting-host=148.87.143.4; posting-account=sIsnLQwAAAARvYJo6PLbFuqdYUfWHtpS Xref: dp-news.maxwell.syr.edu comp.databases.oracle.server:262712 Mladen Gogala wrote: > On Fri, 03 Mar 2006 11:06:00 +0100, Holger Baer wrote: > > > According to the expert, no undo will be generated *IF USING NOLOGGING* option. > > Try to read and understand what you cite. > > > > Regards, > > Holger > that is not what I've said - insert /*+ APPEND */ will bypass UNDO generation for the table data (but not any indexes on said table). It will bypass redo generation for the table (but never the indexes) IF the database is noarchivelog OR nologging has been set on the table, else full logging is performed for the table data (then there is that force logging that can affect this as well at the database level) undo to protect the dictionary (and redo) entries that are made to advance the high water mark will of course be generated. > The trick with the direct insert and undo is that the "old value" doesn't > exist, so there will not be massive undo generation, period. As for > calling upon the authority of experts, I disagree with that as a matter of > principle. I suspect that Tom Kyte also may be proven wrong from time to indeed. > time as he's probably human, just like the rest of us. Tom Kyte is a great I am. One of my mantras is "Question Authority", use it in most all of my seminars as the THEME in fact. > and extremely knowledgeable guy, but checking works best for me. This is > nothing personal: I even check things that Jonathan Lewis says, although > he is definitely more then a mere human. hmmm ;) > > > -- > http://www.mgogala.com