Re: Recovery Question.

From: Niklas Iveslatt <"Niklas>
Date: Tue, 25 Jul 2023 08:53:59 -0600
Message-ID: <CAHLzPNfszjYYmyzqF4TjmdE7usPG8Q6Ah-1diKnHJzKTKVrQbQ_at_mail.gmail.com>



You can set:

_allow_resetlogs_corruption=true
undo_management=manual

and then startup force

But you should do this on a copy of that database (or do a full cold backup) - once you open the database with that option it is irreversible so you want to preserve what you have while you work with Oracle support. If it opens and sort of functions try to do a full export and hope you can get as much as possible and then create a new database and do import. No guarantees here and this is at your own risk.

Good luck!

Niklas Iveslatt
Senior Partner

Arisant LLC ~ http://www.arisant.com
44 Inverness Dr. E Bldg. C Suite 2 ~ Englewood, CO 80112 mobile: 303.882.4461 ~ main: 303.330.4065 ~ fax: 888.889.0155

  Need to send me something securely? *Click here* <https://arisant.sendsafely.com/u/niklas.iveslatt>

On Tue, Jul 25, 2023 at 8:27 AM Peter Gram <peter.m.gram_at_gmail.com> wrote:

> Hi
>
> You need to open at SR with Oracle support to get the underscore parameter
> to set before it is possible to open a database with inconsistent data
> files.
>
> Gram/
>
> On Tue, 25 Jul 2023 at 05.08, Satalabaha Oracle <
> satalabaha.oracle_at_gmail.com> wrote:
>
>> Hi All,
>>
>>
>> I have an Oracle recovery related question..
>>
>> Environment: Oracle 12.1 running on RHEL 7.
>>
>> We performed incomplete recovery for a set of datafiles (646 datafiles)
>> as we lost the mount point that had those datafiles. After the recovery,
>> below is the status of checkpoint for both datafile header and checkpoint
>> information in controlfile. If we notice the checkpoint_change# for all the
>> datafiles are in consistent state and checkpoint_change_change# of datafile
>> header matches with the controlfile too. But Oracle was not allowing to
>> open the database [1] using resetlogs as the SYSTEM datafile requires more
>> recovery.
>>
>> Q1) Why would Oracle still report FUZZINESS for the rest of the 2635 odd
>> datafiles?
>> Q2) What other conditions should be met before we can make sure that the
>> database can be opened?
>>
>> [1]
>> ORA-01547: warning: RECOVER succeeded but OPEN RESETLOGS would get error
>> below
>> ORA-01196: file 1 is inconsistent due to a failed media recovery session
>> ORA-01110: data file 1: '/u01/xxxx/xxxxx/xxxxx_system01.dbf'
>>
>> Details:
>> ============
>>
>> SQL> ALTER SESSION SET NLS_DATE_FORMAT='DD-MON-RR HH24:MI:SS';
>>
>> Session altered.
>>
>> SQL> select checkpoint_change#,checkpoint_time,fuzzy,count(1) from
>> v$datafile_header group by checkpoint_change#,fuzzy,checkpoint_time;
>>
>> CHECKPOINT_CHANGE# CHECKPOINT_TIME FUZZY COUNT(1)
>> -------------------- --------------------------- ---------
>> --------------------
>> 48336396202 28-JAN-23 11:02:52 NO 2
>> 78233391650 23-JUL-23 08:57:11 NO 646
>> 78233391650 23-JUL-23 08:57:11 YES 2635
>>
>> SQL> select checkpoint_change#,checkpoint_time,count(1) from v$datafile
>> group by checkpoint_change#,checkpoint_time;
>>
>> CHECKPOINT_CHANGE# CHECKPOINT_TIME COUNT(1)
>> -------------------- --------------------------- --------------------
>> 48336396202 28-JAN-23 11:02:52 2
>> 78233391650 23-JUL-23 08:57:11 3281
>>
>>
>> --
>> Thanks,
>> Satalabaha
>>
> --
> Best regards/Venlig hilsen
>
> Peter Gram
> Sæbyholmsvej 18 DK-2500 Valby
>
> Mobile: (+45) 5374 7107
> Email: peter.m.gram_at_gmail.com
>
> <http://oaktable.net/members>
>

--
http://www.freelists.org/webpage/oracle-l
Received on Tue Jul 25 2023 - 16:53:59 CEST

Original text of this message