1 Reply Latest reply: Jan 15, 2014 5:37 AM by mseberg RSS

    Switchovers in ROLE TRANSACTIONS OF DATA GUARD

    user9085814

      Hi All,

       

      I am performing a switchover operation from PRIMARY to STANDBY database .

      From the alert log file i can see 3 log switches are happening during switchover operation.

       

      Please find the alert log below

       

      Destination LOG_ARCHIVE_DEST_2 is SYNCHRONIZED

      ******************************************************************

      LGWR: Setting 'active' archival for destination LOG_ARCHIVE_DEST_1

      ******************************************************************

      LGWR: Standby redo logfile selected for thread 1 sequence 24 for destination LOG_ARCHIVE_DEST_2

      Thread 1 advanced to log sequence 24

        Current log# 3 seq# 24 mem# 0: /u01/PRD/redo03.log

      Fri Jan 03 08:11:32 2014

      Starting background process CJQ0

      Fri Jan 03 08:11:32 2014

      CJQ0 started with pid=22, OS id=7884

      Fri Jan 03 08:12:22 2014

      alter database commit to switchover to physical standby

      ALTER DATABASE COMMIT TO SWITCHOVER TO PHYSICAL STANDBY (PRD)

      Fri Jan 03 08:12:24 2014

      LGWR: Standby redo logfile selected for thread 1 sequence 25 for destination LOG_ARCHIVE_DEST_2

      Thread 1 advanced to log sequence 25

        Current log# 1 seq# 25 mem# 0: /u01/PRD/redo01.log

      Waiting for all non-current ORLs to be archived...

      ********************************************************************

      LGWR: Resetting 'active' archival for destination LOG_ARCHIVE_DEST_2

      ********************************************************************

      ********************************************************************

      LGWR: Resetting 'active' archival for destination LOG_ARCHIVE_DEST_1

      ********************************************************************

      Destination LOG_ARCHIVE_DEST_2 no longer supports SYNCHRONIZATION

      Thread 1 advanced to log sequence 26

        Current log# 2 seq# 26 mem# 0: /u01/PRD/redo02.log

      Fri Jan 03 08:12:29 2014

      Stopping background process CJQ0

      Fri Jan 03 08:12:29 2014

      SMON: disabling tx recovery

      Fri Jan 03 08:12:32 2014

      Stopping background process QMNC

      Stopping Job queue slave processes, flags = 27

      Waiting for Job queue slaves to complete

      Job queue slave processes stopped

      Stopping background process FBDA

      SMON: disabling cache recovery

      Fri Jan 03 08:12:37 2014

      Shutting down archive processes

      Archiving is disabled

      Fri Jan 03 08:12:37 2014

      ARC3: Becoming the heartbeat ARCH

      ARC3: Archiving disabled

      ARCH shutting down

      ARC3: Archival stopped

      Fri Jan 03 08:12:37 2014

      ARCH shutting down

      ARC2: Archival stopped

      Fri Jan 03 08:12:37 2014

      ARCH shutting down

      ARC1: Archival stopped

      Fri Jan 03 08:12:37 2014

      ARCH shutting down

      ARC0: Archival stopped

      Thread 1 closed at log sequence 26

      Successful close of redo thread 1

      ARCH: Noswitch archival of thread 1, sequence 26

      ARCH: End-Of-Redo Branch archival of thread 1 sequence 26

      ARCH: Archiving is disabled due to current logfile archival

      Using STANDBY_ARCHIVE_DEST parameter default value as /u01/PRD/arch

      LOG_ARCHIVE_DEST_2 is a potential switchover target

      Backup controlfile written to trace file /apps/oracle/diag/rdbms/prd/PRD/trace/PRD_ora_7841.trc

      Clearing standby activation ID 1899035605 (0x7130fbd5)

      The primary database controlfile was created using the

      'MAXLOGFILES 16' clause.

      There is space for up to 13 standby redo logfiles

      Use the following SQL commands on the standby database to create

      standby redo logfiles that match the primary database:

      ALTER DATABASE ADD STANDBY LOGFILE 'srl1.f' SIZE 52428800;

      ALTER DATABASE ADD STANDBY LOGFILE 'srl2.f' SIZE 52428800;

      ALTER DATABASE ADD STANDBY LOGFILE 'srl3.f' SIZE 52428800;

      ALTER DATABASE ADD STANDBY LOGFILE 'srl4.f' SIZE 52428800;

      Archivelog for thread 1 sequence 26 required for standby recovery

      Fri Jan 03 08:12:39 2014

      MRP0 started with pid=17, OS id=7904

      MRP0: Background Managed Standby Recovery process started (PRD)

      Fast Parallel Media Recovery NOT enabled

      Managed Standby Recovery not using Real Time Apply

      parallel recovery started with 2 processes

      Online logfile pre-clearing operation disabled by switchover

      Media Recovery Log /u01/PRD/arch/arch_1_26_835358680.arch

      Identified End-Of-Redo for thread 1 sequence 26

      Resetting standby activation ID 1899035605 (0x7130fbd5)

      Media Recovery End-Of-Redo indicator encountered

      Media Recovery Applied until change 542157

      MRP0: Media Recovery Complete: End-Of-REDO (PRD)

      MRP0: Background Media Recovery process shutdown (PRD)

      Fri Jan 03 08:12:45 2014

      Switchover: Complete - Database shutdown required (PRD)

      Completed: alter database commit to switchover to physical standby

       

      My question here is Switchover operation is performing  3 log switches

      > ONE FOR sequence number 24 -- This is for archiving the online redo logs before switchover operation

      > ONE FOR sequence number 25 -

      > ONE FOR sequence number 26 - This is for Marking of the End of Redo log for the current activation id

       

      Is there any specific reason for having an extra log file operation [here sequence number 25] between the switchover operation and END OF REDO Log operation ?

      Is it an internal behaviour ?

       

       

      Thanks

        • 1. Re: Switchovers in ROLE TRANSACTIONS OF DATA GUARD
          mseberg

          Hello;

           

          I'm not sure I read it that way. You could look at it as 25 is the current redo and 26 is the last. Since 24 was before the switchover why would it count?

           

          I would probably compare the file sizes of 25 and 26 to the average size before the switchover for some insight into what's going on. But any Oracle design information on this I cannot speak to the question.

           

          What problem with switchover are you trying to solve?

           

          Best Regards

           

          mseberg