Ora8 NT problem

I ran into a problem on development box at home a month ago. Hopefully someone here can tell me what I did wrong?

I had a bootdisk crash. After getting a new disk and re-installing NT 4 and Oracle 8, I first did a full backup of all the files of my Ora8 db (the datafiles, controlfiles, redo and archive logs were not on the crashed bootdisk) to another drive.

Next I tried to get the database up - only problem was that with my installing the new bootdisk and NT, the rest of the disks were now named differently. The Oracle datafiles which used to be on drive E: and were now on drive F:

Not a problem I though, re-create the control file. So I recreated the control file and tried opening the database. Then Oracle complained that the system tablespace was stuffed and needed to be restored. So I tried to recover it, but no luck - it was looking for log 145 and my last log was only at 144!

After a lot of struggling I decided to hell with this and rename the drive letter for the data disk back to E: And then without having to do anything special, the database came up with a normal startup command.

The alert logfile also contained some funnies I though. When I tried to bring up the database on drive F: with the new controlfile, the alertlog said that -all- the datafiles were fuzzy. And of course, server manager insisted on the system tablespace being recovered. Why is this? I was under the impression that when you re-create a control file, it will be in sync with the datafiles. And there was nothing wrong with the datafiles as the db came up without any hassles when I changed drive F: back to E:

I can re-create the whole problem and then try it again setting the trace levels on for dumping the file headers of the datafiles and look at the SCN's and other weird stuff, but that is a bit extreme - surely an Ora8 db should easily be recoverable in this situation.

Any idea where I went wrong?

regards,
Billy Received on Fri Mar 06 1998 - 00:00:00 CST