Databases
cancel
Showing results for 
Search instead for 
Did you mean: 

What will happen if redolog file is cleared in oracle8i?

Pradeep_3
Frequent Advisor

What will happen if redolog file is cleared in oracle8i?

In case of redolog file (without member) corruption, What i need to do?

If i am not wrong i need to clear the redolog file.

If i clear the redolog file, Whether oracle will allocate new log sequence number or it will use the existing redolog sequence number.
Redolog file will have both commited and uncommited data.So what will happen to this data??
Also i have gone through the documentation, they have mention the in case of clearing redolog file. It will invalidate the standby site?
I am not able to understand this point. Why it will invalidate my standby database?

Pls. Any one can explain me this point????????
6 REPLIES
malay boy
Trusted Contributor

Re: What will happen if redolog file is cleared in oracle8i?

Redolog file corrupt no member ,then you are in deep problem.

This what you do :

1) shutdown the database and mount it again.
2) do :
alter database clear unarchived logfile group
3) you have to do incomplete recovery.
do:
select * from v$log;

look for column seq.

4) restore from backup.
5) do :

svrmgrl > recover database until cancel;
put in the last sequence number .
6) alter database open resetlogs;
7) pray to god that above step are OK.


Simple advice,please-please duplex your redolog file.

The above step are taken from "DATABASE ADMIN FUNDAMENTAL II" under subject loss of current redo log files.

regards
mB

There are three person in my team-Me ,myself and I.
twang
Honored Contributor

Re: What will happen if redolog file is cleared in oracle8i?

1. First, try to clean up any client connections by killing them at the
operating system level.

2. You cannot open the database as you again receive the same errors on the
corrupted redo log. You must bring the database to the mount stage and
clear the logfile with:

ALTER DATABASE CLEAR UNARCHIVED LOGFILE '/oradata/redo_1.log'

3. You are then able to open the database and do a clean shutdown.

4. At this point, you should take a backup or replace the failing hardware.

Pradeep_3
Frequent Advisor

Re: What will happen if redolog file is cleared in oracle8i?

Dear Twang

Thanks for quick reply but I am aware of steps how to do it. My question is why will it invalidate standby database.

Massimo Bianchi
Honored Contributor

Re: What will happen if redolog file is cleared in oracle8i?

The standby db will be indalidated because of the steps involved.

Every recovery invalidate the standby database, because SCN are going back.

The corruption the same: archive are transferred after the redolog is completed, not before, so you would have incomplete information to transmit (and are not transferred!).

But, in that case, you have the (hopefully good) standby database.

It's a theoric question or a pratical question ?

If theoric, take the above good advice and have at least two member on ohisically separate device for each log, if much better!

Massimo
twang
Honored Contributor

Re: What will happen if redolog file is cleared in oracle8i?

If you clear log files at the primary database by issuing:
alter database clear unarchived logfile;
or open the primary database using the resetlogs option, you invalidate the standby database.
The first command clears the online logfile (fills it with all null). This makes the corruption disappear which causes the contents of the cleared online redo file to be missing in the archived redo stream.
Yes, I agree that you must multiplex the online redo logs in the db.
If you have added redo logs on the primary database, you need to refresh the standby controlfile. Additionally, if the directory location/name is different, you will need to reflect that in the LOG_FILE_NAME_CONVERT parameter.
R. Allan Hicks
Trusted Contributor

Re: What will happen if redolog file is cleared in oracle8i?

It will invalidate the standby for the same reason that it invalidates the on-line system, and that is that no archive log can be generated from the bad archive log.

The other respondents have given you good advice on the recovery. I would like to add, that if you generate a check point so that the datafiles will be updated from memory and the redo's will not be neccessary for re-start. DO a clean shutdown and bring the database back up. A point in time recovery may be necessary, but the checkpoint gives you 1/2 a chance of not having to roll any changes out of the redologs. Then, go through the procedure to initialize your standby database again.

I'm not that familar with how 8i does the hot standby, but I thought that it synchronized the databases via the archivelogs. Hence, no archive log because the redo can't be archived, leads to a dead standby.

Dataguard under 9i allows the logwriter to write to the standby redologs. I've played with dataguard a little and like it. It doesn't like datafiles above 2gbyte, but if you install the latest perl it will work.


BTW There is/was a bug in 8i (I forget the rev 8.1.6 I think) that if you have a bad member file in the redo group. The database will give you three warnings that there is a bad member and then do a polite shutdown.
"Only he who attempts the absurd is capable of achieving the impossible