Re: ORA-01113_ file 1 needs media recovery
Date: Mon, 27 Apr 2009 05:19:27 -0700 (PDT)
Message-ID: <0236d336-b6ae-4a82-8ec5-235141b455b7_at_w31g2000prd.googlegroups.com>
On Apr 26, 8:17 pm, Gassamaba <ogs..._at_gmail.com> wrote:
snip
> Hi Austin,
> Yes, the database isd too old and we are in the process of migrating
> to a different system and just happened to be down before we could
> extract the required information (whole content of four data tables)
> for the new system.
> Anyway, we had two contoel files on this system. the startup process
> recognise only one control file and we managed to bypass the contol
> file error by copying the recognised control file into the raid5
> folder then edit 'init.ora' file.
>
> Is there any danger to force database startup so as to enable us
> exract/dump content of certain data tables? How?
> Any suggestions will be usefule.
Please stop top posting most people on cdos will not respond to top posters.
If I were you I would get a backup of how the database is now before proceeding any further in your experiments.
Back to your specific question it is a tried and proven method of copying a good control file over the top of a bad alternate control file. You want to have the instance down or at least not mounted when you do a copy like that then try an alter database mount.
Was your system in archivelog mode? Do you have archivelogs thru the tape backup you are attempting to recover from? What kind of tape backup was it ... cold ... hot ... ???
The message that your system datafile needs further recovery is not good ... usually you would have to be in a position where oracle has already satisfied itself about the control file(s) it is using before any kind of message about other datafiles would come out. Received on Mon Apr 27 2009 - 07:19:27 CDT