yesterday with a datafile being corrupted.
I performed an RMAN restore, after double-checking the datafile was
offline, with:
restore datafile 32;
It ran for a while, but then bombed out with the following:
channel ORA_DISK_1: restored backup piece 14
piece handle=E:\BACKUP\BACKED UP DATA\AJIGLMEM_14_1.BAK
tag=TAG20070501T220157
channel ORA_DISK_1: reading from backup piece E:\BACKUP\BACKED UP
DATA\AJIGLMEM_15_1.BAK
ORA-19870: error reading backup piece E:\BACKUP\BACKED UP
DATA\AJIGLMEM_15_1.BAK
ORA-19504: failed to create file "E:\ORACLE\DATAFILES\NGDB_DATA_06.DBF"
ORA-27086: unable to lock file - already in use
OSD-00002: additional error information
O/S-Error: (OS 32) The process cannot access the file because it is being
used b
y another process.
failover to previous backup
creating datafile fno=32 name=E:\ORACLE\DATAFILES\NGDB_DATA_06.DBF
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 05/03/2007 14:28:54
ORA-01119: error in creating database file
'E:\ORACLE\DATAFILES\NGDB_DATA_06.DBF'
ORA-27038: created file already exists
OSD-04010: <create> option specified, file already exists
This Oracle 10.2.0.1 on a windows 2003 Server machine.
Would (could) I just take the entire tablespace offline, rename the
existing datafile, and then re-run the restore? Or stop Oracle instead to
rename the file, re-run the restore, then restart Oracle? Or am I missing
something else here?
This is actually the first time I've ever had to perform a restore in 18
years so I'm a bit gun-shy, though it wouldn't be the end of the world if I
had to drop the tablepace and recreate it, I'd only lose abround a weeks
worth of work, but no big deal.
Thanks.
--
-- Bill Ferguson
