Hey Paul, I think you missed the part about this being an *upgrade* to 9i! ;) 10g unfortunately isn't an option. There were a few (apparently) heated calls with the vendor about us even going to 9i. Your logs look the same as most of our Unix ones. I'm guessing that the timestamps are put there by the ARCH process then, because our logs have none. Thanks for the reply! Rich -----Original Message----- From: Paul Drake [mailto:bdbafh@xxxxxxxxx] Sent: Tuesday, March 14, 2006 2:14 PM Subject: Re: No timestamps for redolog switches on Windohs? Rich, Sorry, but I don't have any 9.2 win32 dbs to test against. Perhaps an upgrade to 10g R1 with 10.1.0.4 and 10.1.0.4 patch 10 might be in order? Perhaps 10.1.0.5 + the CPUJan2006 patch instead? Here is a snipped from an alert log from yesterday. Yes, the online redo logs are being switched here far too frequently. No, the online redo logs are sized just fine - there's just too much redo being generated. That is another matter. Paul Mon Mar 13 17:22:43 2006 ARC1: Evaluating archive thread 1 sequence 4911 Committing creation of archivelog 'H:\ORACLE\ORADATA\MYDB\ARCHIVE\MYDB_562170012_1_4911.LOG' Mon Mar 13 17:23:03 2006 Private_strands 18 at log switch Thread 1 advanced to log sequence 4913 Current log# 1 seq# 4913 mem# 0: E:\ORACLE\ORADATA\MYDB\REDO01A.LOG Mon Mar 13 17:23:03 2006 ARC1: Evaluating archive thread 1 sequence 4912 -- //www.freelists.org/webpage/oracle-l