Re: RMAN Duplicate DB between servers
Date: Wed, 5 Dec 2012 12:04:52 +0800
Message-ID: <CAPjcp6Bp4gZU4eDCymRVPkEF0R+dC1rRDq=Utvcm0qHOPAHmzw_at_mail.gmail.com>
Hi Steve,
Have you tried the steps in this metalink doc: How to Configure RMAN to Write to Shared Drives on Windows NT/2000/2003 [ID 145843.1] ?
I know the article title states it's for writing to shared drives, and the error you're encountering is from reading, but I seem to recall a similar instance where this happened to me and I followed the same steps to get it working.
As far as your second question, I've been wondering about that myself as well :) From my experience on trying this on Unix based systems, it doesn't seem to be consistent. Sometimes it makes a difference (faster) and sometimes it averages the same amount of time if I were to manually copy and restore the files myself.
HTH,
OJ
On Wed, Dec 5, 2012 at 10:29 AM, Steve Wales <sjwales_at_comcast.net> wrote:
> Environment: Windows Server 2008 R2 64-Bit / Oracle 11.2.0.3 64 bit
>
> I have a new environment I'm setting up and testing for an upcoming go
> live.
>
> I have a database PROD on node1 which backs up to i:\backups\PROD
>
> I have my controlfile autobackups going to i:\backups\PROD\ctl
>
> PROD is backing up just fine, I have all sorts of files on i: for backups.
>
> I want to test doing a refresh of the DEV database.
>
> DEV exists on NODE2.
>
> Being what I thought was clever, I mapped the I: drive on NODE2 to
> \\NODE1\I$
>
> I went over to NODE1 and shared I:\BACKUPS with my domain account.
>
> I went over to NODE2 and as far as I can tell, I can see everything.
>
> So, I start a CMD window on NODE2, set my environment variables, stop DEV
> and restart it NOMOUNT with a special PFILE I keep on hand that includes
> db_file_name_convert and log_file_name_convert parameters.
>
> Then I do my RMAN block:
>
> run
> {
> allocate auxiliary channel aux1 device type disk;
> set until time "to_date('2012-12-04:15:00:00','YYYY-MM-DD:HH24:MI:SS')";
> duplicate target database to DEV;
> }
>
> RMAN starts doing its thing. It identifies the correct controlfile backup
> from the Prod catalog but when it tries to open it, I'm getting ORA-27041
> unable to open file.
>
> Going back to the CMD shell, I can set my current directory to
> i:\backups\PROD\ctl, do a directory listing and see the file there from
> NODE2.
>
> I'm sure that there's some funky Windows security thing that I'm missing
> here. Could it be that even though my CMD shell is running under my domain
> account, the ORACLE service runs under Local System and thus can't see the
> network resource ?
>
> I know I could copy the backup files to a local resource and then do
> "duplicate target database to elldev backup location 'd:\some_other_dir';"
> but I'd like to see if I can work out how to handle it this way.
>
> I suppose while I'm at it I'd also like to solicit comments on which is
> likely to be less network intensive - having RMAN pull the backup files
> over
> the network or having me do it manually and restoring from local copies.
>
> Thanks for any input.
>
> Steve
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Dec 05 2012 - 05:04:52 CET