> I have used "diff" on unix to compare the control files from "cp" method and > "ALTER DATABASE BACKUP CONTROLFILE TO" and found they were different (BTW, > no database structure changes were made during this whole process). Starting from 8i, the incremental checkpoint feature makes CKPT process to update controlfile every 3 seconds anyway (in addition to possible redolog record updates during logswitch and updates during full checkpoint etc). > I am curious and would like to know: In what situation "cp controlfile" > method would prevent me from doing a "COMPLETE RECOVERY"? Would a db > structure change (during hot backup) make "cp controlfile" method not > working? In the situation where Oracle simply says that your controlfile is corrupt and it can't open it. Btw, Oracle controlfiles store old-images of the blocks in them for crash recovery, that way they survive from regular instance failures, but I definitely don't recommend backing up a mounted controlfile by OS utilities (a complete atomic snapshot of the whole database being a rare exception). Tanel. ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------