Re: Create standby for oracle 9i database

Updating how we moved forward.

Following the recommendations from Stromasys, we shutdown the databases and all apps on the server. We then did a dump/copy/restore to get everything up to the Stromasys EC2 in AWS. It worked very well for us, albeit it was slow due to the older hardware on the source side.

Thank you for your suggestions and comments.

On Tue, Dec 20, 2022 at 1:52 PM Sandra Becker <sbecker6925_at_gmail.com> wrote:

> I'll have to get with our Slalom contractors to find out. We don't use an
> rman catalog, but I think I could set one up if needed.
>
> Thank you,
>
> Sandy
>
> On Tue, Dec 20, 2022 at 9:40 AM Peter Hitchman <pjhoraclel_at_gmail.com>
> wrote:
>
>> Hi
>> Tiago has made a very important point, you need to check this out. I
>> checked out Solaris on AWS since I have never seen it and found an AWS post
>> about a Solaris emularor provided by Stromasy, but if the byte ordering of
>> the source andt target systems differ you will need to use the rman convert
>> command. That means you need to be using a rman catalog to do the restore.
>> Also if the byte odering is different I do not think you can create a
>> standby that will be able to apply archived log.
>>
>> Regards
>> Pete
>>
>> On Tue, 20 Dec 2022 at 16:04, Sandra Becker <sbecker6925_at_gmail.com>
>> wrote:
>>
>>> Not really sure what's underneath other than both are Sun Solaris. I
>>> was told they were identical, just one on-prem (over 10 years old) and the
>>> other on AWS. We can't patch since we no longer have Oracle support.
>>>
>>> Sandy
>>>
>>> On Tue, Dec 20, 2022 at 8:06 AM Tholiveira <tholiveira_at_gmail.com> wrote:
>>>
>>>>
>>>>
>>>> Just a tough, do u have different CPU’s architecture on new DC?
>>>> In my case i had to patch to 9.2.0.8 so standby would work even with
>>>> both OS’s were the same HPux versions , but Prod at time was PA Risc and
>>>> standby server was newer Itanium
>>>>
>>>> Br
>>>>
>>>> Tiago
>>>>
>>>> On Dec 20, 2022, at 9:41 AM, Sandra Becker <sbecker6925_at_gmail.com>
>>>> wrote:
>>>>
>>>> 
>>>> I had not checked for read only tablespaces. I will do that today.
>>>> Thank you.
>>>>
>>>> Sandy
>>>>
>>>> On Tue, Dec 20, 2022 at 4:54 AM Peter Hitchman <pjhoraclel_at_gmail.com>
>>>> wrote:
>>>>
>>>>> Hi
>>>>> Are any of the tablespaces read only? That can cause problems with
>>>>> duplication.
>>>>>
>>>>> Regards
>>>>> Pete
>>>>>
>>>>> On Tue, 20 Dec 2022 at 11:38, Frank Gordon <frankagordon_at_gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Hello,
>>>>>> Are there any additional clues in the alert.log or trace file
>>>>>> directories?
>>>>>>
>>>>>> Besides the "better" hardware on the standby server, how else does
>>>>>> the standby server differ from the source database? O/S? Patch level?
>>>>>>
>>>>>> Regards,
>>>>>> Frank
>>>>>>
>>>>>>
>>>>>> On Mon, Dec 19, 2022 at 10:53 PM Sandra Becker <sbecker6925_at_gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> OS - Sun Solaris
>>>>>>> Oracle: EE 9.2.0.5.0
>>>>>>>
>>>>>>> We cannot upgrade this database due to the application it supports
>>>>>>> not working with higher versions of oracle. (yeah, I know, really stupid)
>>>>>>>
>>>>>>> We want to create a standby database on a new server with much
>>>>>>> better hardware. I'm getting an error on the duplicate that I just haven't
>>>>>>> been able to figure out after a couple of days of searching. Everything
>>>>>>> I've read points to duplicate filenames, but I have run all the suggested
>>>>>>> queries and there are no duplicate filenames. We are using the same
>>>>>>> directory structure, so I didn't use db_file_name_convert or
>>>>>>> log_file_name_convert. I verified the "standby" can access the backups of
>>>>>>> the primary database.
>>>>>>>
>>>>>>> I did manage to restore the controlfiles manually from the snap_cf
>>>>>>> file that I copied from the target server to the auxiliary server to see if
>>>>>>> that was where it was failing. Then when I tried to restore the database
>>>>>>> it throws the same error as running it from a script. The server is too
>>>>>>> slow for me to restore manually, my session would be killed long before it
>>>>>>> would finish.
>>>>>>>
>>>>>>> I would really appreciate any help you can provide or point me to a
>>>>>>> useful document/website. Thank you,
>>>>>>>
>>>>>>> Sandy
>>>>>>>
>>>>>>> *Duplication Error*
>>>>>>> [image: image.png]
>>>>>>>
>>>>>>>
>>>>>>> *Duplicate Script*
>>>>>>> sqlplus -S "/ as sysdba" << EOF >> $LOG_FILE
>>>>>>> shutdown immediate
>>>>>>> startup nomount
>>>>>>> EOF
>>>>>>>
>>>>>>> rman target sys/pwd_at_primary auxiliary sys/pwd_at_stdby_db << EOF >>
>>>>>>> $LOG_FILE
>>>>>>> run
>>>>>>> {
>>>>>>> allocate auxiliary channel a1 type disk;
>>>>>>> allocate auxiliary channel a2 type disk;
>>>>>>> allocate auxiliary channel a3 type disk;
>>>>>>> duplicate target database for standby nofilenamecheck dorecover;
>>>>>>> release channel a1;
>>>>>>> release channel a2;
>>>>>>> release channel a3;
>>>>>>> }
>>>>>>> EOF
>>>>>>>
>>>>>>> --
>>>>>>> Sandy B.
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> +353-86-0695383
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Regards
>>>>>
>>>>> Pete
>>>>>
>>>>
>>>>
>>>> --
>>>> Sandy B.
>>>>
>>>>
>>>
>>> --
>>> Sandy B.
>>>
>>>
>>
>> --
>> Regards
>>
>> Pete
>>
>
>
> --
> Sandy B.
>
>

-- 
Sandy B.





--
http://www.freelists.org/webpage/oracle-l




image/png attachment: image.png