From Dennis.Meng@cardinal.com Wed, 25 Jun 2003 12:45:15 -0700 From: "Meng, Dennis" Date: Wed, 25 Jun 2003 12:45:15 -0700 Subject: RE: Statistics problem Message-ID: MIME-Version: 1.0 Content-Type: text/plain Title: RE: Statistics problem Thanks all for the input. Unfortunately we don't have an cloned system to work with and this is version 8.1.7.4 on Tru64. Looks like I will just have to sample more rows and see what happens.   Dennis -----Original Message-----From: Jamadagni, Rajendra [mailto:[EMAIL PROTECTED]Sent: Wednesday, June 25, 2003 12:45 PMTo: Multiple recipients of list ORACLE-LSubject: RE: Statistics problem Dennis, I don't think so, but I have (in 9202) successfully exported stats to a table, re-computed the stats, proving that stats-are-not-the-problem-but-bad-code-is and restored the stats back. What version are we talking here? Do you have a test DB that you can work with? Raj -------------------------------------------------------------------------------- Rajendra dot Jamadagni at nospamespn dot com All Views expressed in this email are strictly personal. QOTD: Any clod can have facts, having an opinion is an art ! -----Original Message----- From: Meng, Dennis [mailto:[EMAIL PROTECTED]] Sent: Wednesday, June 25, 2003 11:15 AM To: Multiple recipients of list ORACLE-L Subject: Statistics problem Hi All, I have a very large database that is experiencing performance problems due to bad query plans. While I suspect the cause is bad statistics ( insufficient number of rows being analyzed ), I don't want to change our analyze scripts without seeing some proof. Is there anyway I can query against the database to show that the statistics are indeed bad? TIA Dennis