Re: Cheap Cheap Database Recovery

From: Michael Austin <maustin_at_firstdbasource.com>
Date: Sun, 22 Feb 2009 14:28:24 -0600
Message-ID: <rziol.5040$PE4.1189_at_nlpi061.nbdc.sbc.com>



pitufo wrote:
> On Feb 21, 8:33 am, pitufo <vandr..._at_gmail.com> wrote:

>> Johnburr: I do a cold backup at 2:30 am. (datafiles, and
>> controlfiles, archivelogs are safe in two different servers)
>> Yes, I am in archive logs
>> I dont another backup at a later time....I only do a
>> 'alter database backup controlfile to 'binary.controlfile.ctl''
>>
>> What I am gaining with that: That's how I can make
>> the database apply the archive logs that were generated after the
>> coldbackup at
>> 2:30 am.
>>
>> Michael: Yes, I will do it right....as I said in my original post,
>> I need to document pros/cons of this method, a standby database, and a
>> RAC solution. Now, this method, stupid or not, it is cheap, and it
>> should work. I don't see in it a reason to fail, Do you see it ?

But since, as you stated, it does not work consistently tells me that it is a train wreck in the making and is therefor in reality NOT a viable solution. It may be "cheap" but if it doesn't work, it is useless.

There are many things you can do with RMAN with regards to recovery.. Spend your time in the RMAN docs and proceed from there - before you corrupt something and have to do a real recovery.

>>
>> Thanks

> 
> Just try to go beyond the reason why I am implementing  this way. This
> is a user-managed incomplete recovery, nothing else, the scripts that
> make it automatic can be ignored. As an incomplete recovery MUST work,
> I don't see why it works sometime and why it fails others.
Received on Sun Feb 22 2009 - 14:28:24 CST

Original text of this message