4 Replies Latest reply: Oct 30, 2009 10:06 AM by Robert Geier RSS

    Archive log is not getting transferred to standby side from primary side

    Mahipal
      Hi All,

      In my data guard setup some times one archive log is not getting transferred to standby side, but the archive logs which generated after that getting transferred to standby side but not applying as they miss the archive sequence.


      I have declared the parameters fal_client and fal_server .i.e. if any archive log is not transferred to standby for any reasons these will take care to transfer the missing archive logs to standby.

      But not sure, why my setup is not transferring some of the archive logs to the standby.

      Can any one please help me on this?

      Thanks,
      Mahi
        • 1. Re: Archive log is not getting transferred to standby side from primary side
          Robert Geier
          Check your alert log for errors on primary and standby. Are you using ARCH or LGWR transport ? Check the trace files for those processes.
          • 2. Re: Archive log is not getting transferred to standby side from primary side
            Fahd.Mirza
            Please provide last lines of your alert log file.

            regards
            • 3. Re: Archive log is not getting transferred to standby side from primary side
              Mahipal
              Hi,

              These are excerpts from the alert log.

              LGWR: Standby redo logfile selected to archive thread 1 sequence 7798
              LGWR: Standby redo logfile selected for thread 1 sequence 7798 for destination LOG_ARCHIVE_DEST_2
              Fri Oct 30 05:28:15 2009
              Thread 1 advanced to log sequence 7798 (LGWR switch)
              Current log# 2 seq# 7798 mem# 0: /oradata/soadp/redo/axprod/redo02.log
              Current log# 2 seq# 7798 mem# 1: /oradata/soadp/redo/axprod/redo2b.log
              Fri Oct 30 06:10:16 2009
              LGWR: Attempting destination LOG_ARCHIVE_DEST_2 network reconnect (272)
              LGWR: Destination LOG_ARCHIVE_DEST_2 network reconnect abandoned
              Fri Oct 30 06:10:16 2009
              Errors in file /oracle/admin/axprod/bdump/axprod1_lgwr_409606.trc:
              ORA-00272: error writing archive log
              LGWR: Network asynch I/O wait error 272 log 2 service '(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=tcp)(HOST=u060soad21)(PORT=1521)))(CONNECT_DATA=(SERVICE
              NAME=axdbdrXPT)(INSTANCE_NAME=axdbdr1)(SERVER=dedicated)))'
              Fri Oct 30 06:10:16 2009
              Destination LOG_ARCHIVE_DEST_2 is UNSYNCHRONIZED
              LGWR: Failed to archive log 2 thread 1 sequence 7798 (272)
              Fri Oct 30 06:10:18 2009
              LGWR: Closing remote archive destination LOG_ARCHIVE_DEST_2: '(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=tcp)(HOST=u060soad21)(PORT=1521)))(CONNECT_DATA=(
              SERVICE_NAME=axdbdr_XPT)(INSTANCE_NAME=axdbdr1)(SERVER=dedicated)))' (error 272)
              (axprod1)
              Fri Oct 30 06:10:18 2009
              Errors in file /oracle/admin/axprod/bdump/axprod1_lgwr_409606.trc:
              ORA-00272: error writing archive log
              LGWR: Error 272 closing archivelog file '(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=tcp)(HOST=u060soad21)(PORT=1521)))(CONNECT_DATA=(SERVICE_NAME=axdbdr_X
              PT)(INSTANCE_NAME=axdbdr1)(SERVER=dedicated)))'
              Fri Oct 30 06:10:23 2009



              It shows that the network issue caused the archive log# 7798 did not transfer to standby side.


              But my question is, after the network issue resolved why the archive log# 7798 did not transfer to standby side, where as the the next generated archive logs transferred.

              Also I declared fal_client and fal_server at the time of setup to fill the gap, but seems they have not taken action to transfer the missing archive log# 7798 to standby.

              Thanks,
              Mahi
              • 4. Re: Archive log is not getting transferred to standby side from primary side
                Robert Geier
                please give the results of :-

                show parameter log_archive_dest
                show parameter fal

                Is the database in managed recovery ? If there is a gap, then managed recover should use the fal_server to fill the gap. If it does not, then there should be other error messgages. I suggest you stop and start managed recovery to see if it fills the gaps, or gives an error.