8 Replies Latest reply: May 20, 2013 10:34 AM by Stefan Koehler RSS

Alert log file - FAL archive failed, see trace file.

Jordan Tander
Currently Being Moderated

Dear all,

 

 

Another error on my alert log file:

 

ARC0: Archive log rejected (thread 1 sequence 91879) at host 'PRD_STANDBY'                                        
FAL[server, ARC0]: FAL archive failed, see trace file.                                                            
ARCH: FAL archive failed. Archiver continuing                                                                     
ORACLE Instance PRD - Archival Error. Archiver continuing. 

 

 

The archivelogs shipping is working fine, just checked if they are automatically applied on standby --- and no issues.

 

On primary, I have:

 

SQL> show parameter FAL_CLIENT;

 

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

fal_client                           string      PRD_PRIMARY

 

 

SQL> show parameter LOG_ARCHIVE_DEST_2;

 

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

log_archive_dest_2                   string      SERVICE=PRD_STANDBY COMPRESSIO

                                                 N=ENABLE optional reopen=60 ma

                                                 x_connections=5

 

 

 

On standby I have:

 

SQL> show parameter FAL_CLIENT;

 

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

fal_client                           string      PRD_STANDBY

 

SQL> show parameter LOG_ARCHIVE_DEST_2;

 

NAME                                 TYPE        VALUE

------------------------------------ ----------- ------------------------------

log_archive_dest_2                   string      SERVICE=PRD_PRIMARY COMPRESSIO

                                                 N=ENABLE optional reopen=60 ma

                                                 x_connections=5

 

 

Thanks in advance for some ideas,

 

J.

  • Re: Alert log file - FAL archive failed, see trace file.
    Deepak Kori
    Currently Being Moderated

    Hi Jordan,

     

    Do we have archiver full situation on PRD instance ?

     

    Regards,

    Deepak Kori

  • Re: Alert log file - FAL archive failed, see trace file.
    Reagan Benjamin
    Currently Being Moderated

    Hello

    It is definitely not an archiving file system full. If it was the case then you would see archiver stuck messages in the alert log.

    This is due to a bug and Oracle suggests to ignore these messages appearing in the Alert log as long as the Primary and Standby are in sync.

     

    RC3: Archive log rejected (thread 1 sequence 136480) at host '(DESCRIPTION=  (ADDRESS=(PROTOCOL=tcp)(HOST=192.168.1.26)(PORT=1530))(CONNECT_DATA=(SERVICE_NAME=sdbinfo)(SERVER=DEDICATED)))'

    Errors in file /opt/dbinfo/diag/rdbms/dbinfo/dbinfo/trace/dbinfo_arc3_7445.trc:

    ORA-16401: archivelog rejected by RFS

    FAL[server, ARC3]: FAL archive failed, see trace file.

    Errors in file /opt/dbinfo/diag/rdbms/dbinfo/dbinfo/trace/dbinfo_arc3_7445.trc:

    ORA-16055: FAL request rejected

    ARCH: FAL archive failed. Archiver continuing

    ORACLE Instance dbinfo - Archival Error. Archiver continuing.

     

    - RedoLog switch happens very frequently before the above Error reported

    - The requested ArchiveLog by FAL requests the current Log Sequence or the Sequence currently being archived

    Changes

     

    The Problem here is that the Primary Database is switching Logs too frequently.

     

    Using ARCH to send the archives, every time there's a log switch the Primary has to send the Archivelog to the Standby, meanwhile another Log Switch occurred on the Primary which causes also another Archivelog to be sent to the Standby, but the first one has not finished yet, a GAP is formed and detected by the Standby. At this Time the first Archivelog is also sent as FAL Request, but this one will fail because the first one is still being archiving, locked, so the second one fails.

     

    Solution

     

        Ignore these Messages as long as the Standby Database keeps synchronized with the Primary

        Database Increase the Size of the Online Redologs to reduce Redolog Switch Frequency

        Increase Network Bandwith between the Primary and Standby Database

     

    Cheers

    RB

Actions