Hi,

I am using Oracle 9i database and trying to do a recovery from my prod
machine (System = SunOS,Release = 5.8,KernelID = Generic_108528-14)
to development machine(GNU/Linux) platform(Linux devdb 2.4.20-8 #1 Thu
Mar 13 17:54:28 EST 2003 i686 i686 i386 GNU/Linux ).

My production database in one machine say prod1 and the catalog of the
daily backup is stored in another machine(same configuration as prod)
prod2 . Now I am trying a procedure which will directly connect rman
in dev machine to prod2 and do the recovery in dev machine. Now all
back-up control files reside in prod1 machine.

First I tried a procedure like this:
I created a new database in dev machine which has same database name
as prod and same directory structure(ie using initprod.ora and
creating same directory sturcture for .log and other datafiles such as
system.dbf)


1.Copy the initialization parameter file for HOST_A to HOST_B using an
operating system utility.
2.Connect to the HOST_B target instance and HOST_A recovery catalog.
ie:
rman target sys/change_on_install catalog rman/xxxxx@backup


3.Start the instance without mounting it:
startup nomount;
4.Restore and mount the control file. Execute a run command with the
following sub-commands:

run {
allocate channel ch1 type disk;
restore controlfile;
alter database mount;
}

5.Because there may be multiple threads of redo, use change-based
recovery. Obtain the SCN for recovery termination by finding the
lowest SCN among the most recent archived redo logs for each thread.
Start SQL*Plus and use the following query to determine the necessary
SCN:
SELECT min(scn)
FROM (SELECT max(next_change#) scn
FROM v$archived_log
GROUP BY thread#);
6.Execute a run command with the following sub-commands:
a.Set the SCN for recovery termination using the value obtained from
the previous step.
b.Allocate at least one channel.
c.Restore the database.
d.Recover the database.
e.Open the database with the RESETLOGS option.
f. run {
g. set until scn = 500; # use appropriate SCN for
incomplete recovery
h. allocate channel ch1 type disk;
i. restore database;
j. recover database;
k. alter database open resetlogs;
l. }


Then we got the Error when we tried to do STEP 4: ie "Restore and
mount the control file"


ORA-19505: failed to identify file
"/backup/controlfile/PROD_516506910_2563_1.cf"
ORA-27037: unable to obtain file status
Linux Error: 2: No such file or directory

So we copied /backup/controlfile/PROD_516506910_2563_1.cf file from
prod1 to dev box under same directory structure and we re run the
procedure

Now we are getting the Error

RMAN-03002: failure of restore command at 01/28/2004 02:02:48
ORA-19587: error occurred reading 0 bytes at block number 1
ORA-27091: skgfqio: unable to queue I/O
ORA-27067: skgfdisp: size of buffer is invalid
Additional information: 2

Could you please help us !

On 28 Jan 2004 02:16:19 -0800, aravind (AT) digiterrabroadband (DOT) com
(Aravind) wrote:

Instead of reinventing the wheel (and crossposting to all Oracle
groups you can spell and annoy people)
why don't you just read the RMAN manual?
It will provide you with all the answers


--
Sybrand Bakker, Senior Oracle DBA

Hi Aravind,

Check if copied controlfile has "read-only" permission.
If so, this might be a problem, change it to "rw" and
try again.

Regards,
Andrew


aravind (AT) digiterrabroadband (DOT) com (Aravind) wrote in message news:<67fd4548.0401280216.1b3b0a65 (AT) posting (DOT) google.com>...

Hi,

possible it will not work because cross-plattform recovery of a
database isn't supported.


regards

s.kapitza


aravind (AT) digiterrabroadband (DOT) com (Aravind) wrote in message news:<67fd4548.0401280216.1b3b0a65 (AT) posting (DOT) google.com>...