Re: Database recovery

Hello world,

[Alex, thanks]

This quote referred to a case where both the actual control files and the
online redo logfiles were lost.
If the online redo logfiles are still there (you only lost the controlfiles,
and maybe some datafiles) then the UNTIL clause is NOT needed since the RECOVER
command will then detect the end of the recovery by itself. When in that case the RECOVER command says it is ready, it means that the database
is in the same state as before the crash.

Erwin
---
Erwin Dondorp
www.dondorp.com

Alex Hudghton wrote:
> To steal a post from Erwin Dondorp...............
> When you use a backup controlfile, you always need to add the clause
> "UNTIL ..."
> Otherwise the rdbms will look for proof that recovery is complete by
> looking at the online redo logfiles
> and/or the controlfiles, but these are already modified/recreated/lost
> in your case.
> I suggest you use "UNTIL CANCEL":
> "RECOVER DATABASE USING BACKUP CONTROLFILE UNTIL CANCEL"
> The RECOVER command is the interactive version of "alter database
> recover".
> >Yes. After I done the backup of datafile, I issued the
> >switch logfile and then do the copy of archive log
> >> Are you doing the switch logfile after you have taken the tablespaces
> >> out of hot backup mode ??
> >> >I am trying out my database recovery procedure. I have
> >> >done a hot backup of the whole database , then switch log
> >> >and backup the archive log. THen I do a backup of the control
> >> >file using oracle svrmgrl.
> >> >After this I erase the whole database and copy back
> >> >all my datafile. I have also copied the control file
> >> >as well. When I try to start the database, it gives
> >> >following:
> >> > recover database using backup controlfile
> >> >ORA-00279: Change 16341 generated at 11/18/99 22:53:43 needed for
> >thread
> >> >1
> >> >ORA-00289: Suggestion : /ora05/oradata/TEMS/arch/arch1_32.dbf
> >> >ORA-00280: Change 16341 for thread 1 is in sequence #32
> >> >Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
> >> >However, I do not have seq 32. What could have gone wrong?
> >> >I have applied seq 31 though as suggestef by Oracle.
Received on Tue Nov 23 1999 - 16:25:21 CST