Re: Snapshot Thin Clone of Physical Standby
Date: Mon, 5 Oct 2015 09:56:16 -0500
Message-ID: <CAHJZqBAYDVadKbtvnexqwCzmNSO0ndzb-rME3yeYv3KFJR63og_at_mail.gmail.com>
Seth, that did appear to work now when I create a new controlfile. I'll see how best to script that part of it.
Thanks much for your help.
Don.
On Sun, Oct 4, 2015 at 5:09 PM, Seth Miller <sethmiller.sm_at_gmail.com> wrote:
> Don,
>
> I'm always creating clones so I just create a new control file. This is
> also how all of the DB cloning products work as well as the E-Biz dbTier
> cloning.
>
> It's ok to continue log shipping since your clone won't need those changes.
>
> Seth Miller
>
> On Sun, Oct 4, 2015 at 7:27 AM, Don Seiler <don_at_seiler.us> wrote:
>
>> Seth & Mark,
>>
>> When you did this, did you create a new control file or use the cloned
>> the standby control file and activate the clone? I did one test with
>> stopping MRP but still got the error about renaming the SRL (testing on
>> same host). Note that log shipping to the original standby was still
>> running.
>>
>> I wonder if we should just activate the cloned standby at this point?
>>
>> Don.
>> On Oct 3, 2015 2:42 PM, "Mark Burgess" <mark_at_burgess-consulting.com.au>
>> wrote:
>>
>>> That is the same approach we use for this - cancel MR on the standby,
>>> snapshot the file systems (ZFS in this case) and then activate the cloned
>>> standby in the target environment which takes care of the redo logs etc.
>>>
>>> Performing this type of cloning is greatly simplified when you have the
>>> option to clone to a different host. Previously when using the same host
>>> the approach then required renaming the control file, redo log, temp etc
>>> along with using RMAN to do the datafile rename (catalog the snapshot
>>> directory, switch to datafile copy).
>>>
>>>
>>> On 4 Oct 2015, at 6:29 AM, Seth Miller <sethmiller.sm_at_gmail.com> wrote:
>>>
>>> I've done this successfully without the need for recovery. If you cancel
>>> managed recovery before taking the snapshot, the standby will do a
>>> checkpoint and give you a consistent database.
>>>
>>> Seth Miller
>>>
>>>
>>>>
>>>
>
-- Don Seiler http://www.seiler.us -- http://www.freelists.org/webpage/oracle-lReceived on Mon Oct 05 2015 - 16:56:16 CEST