cancel
Showing results for 
Search instead for 
Did you mean: 

Redo log - Standby database

Former Member
0 Kudos

Hello,

Oracle 11g .

We have 2 physical standby's for our Production database :  A & B at different locations.

On A : Oracle data guard is used for logs shipping and applying logs .

On B : Orcale data guard is used only for logs application and shipping is done via SFTP from A.

Frequently we have started to receive daily alerts for standby B like:

The Following ORACLE Errors have been detected in

/oracle/SID/saptrace/background/alert_SID.log   at approximately   09:47:00 (PST) on 03/21/13

ORA-308 signalled during: alter database register logfile '/oracle/SID/saparch/SIDarch1_2083967_624808420.dbf'...

----

I am very new in DBA , so would like to know how should i try to investigate this issue.

Thanks.

Accepted Solutions (0)

Answers (1)

Answers (1)

stefan_koehler
Active Contributor
0 Kudos

Hi,

> I am very new in DBA , so would like to know how should i try to investigate this issue.

.. the only wise reply is "Get some training first and the needed skill set from an experienced DBA or consultant", seriously. It seems like you are running a 3 node Data Guard environment without the basic knowledge at all. Data Guard is usually used in mission critical environments and the guys should know what they are doing in such cases.

Why do you ship the redo logs by SFTP at all? You are running an Oracle EE, if this is an Oracle database for SAP and so you can use all the integrated features for sending and receiving redo streams / logs. Security concerns are also not a valid reason as you can tunnel the redo stream by SSH.

However i think the ORA-308 error is pretty self-explanatory.

shell> oerr ORA 308

00308, 00000, "cannot open archived log '%s'"

// *Cause:  The system cannot access a required archived redo log file.

// *Action: Check that the off line log exists, the storage device is

//          online, and the archived file is in the correct location.

//          Then attempt to continue recovery or restart the recovery

//          session.

There are special cases / issues for logical standby databases and dNFS, but the first is not supported at all in a SAP environment and the second case is not mentioned.

Regards

Stefan