Frequent ORACLE Database restart failure caused by cold Backup Recovery

Environment: ORACLE 11G R2 + windows2012

Description: After the customer restores the cold backup to start the database, about 30 seconds after the start of the database can not be operated, the prompt ORA-03113: The end of the communication channel file solution, it takes about 30 seconds to restart.

Processing Process:

1. Go to the operating system to view related logs.

View log error, found ORA-00600: internal error code, arguments: [3005]

At the same time, it is observed that the time of the undo tablespace physical file in the DATA directory has not changed, and other physical file updates have become the latest time.

Therefore, the physical files in the redo tablespace are rebuilt.

The procedure is as follows:

Sqlplus/nolog

SQL> conn/as sysdba;

SQL> startup mount;

SQL> -- create a static parameter file based on the current dynamic parameter file

SQL> create pfile = 'd: \ 20130101. ora 'from spfile;

# Modify the generated 20130101. ora File

*. Undo_management = 'manual'

*. _ Uploupted_rollback_segments = (_ SYSSMU3 $)

[Oracle @ DB ~] $ Exit

SQL> -- view the name of the original undo tablespace

SQL> show parameter undo;

NAME TYPE VALUE

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

Undo_management string AUTO

Undo_retention integer 10800

Undo_tablespace string UNDOTBS1

SQL> -- close the database

SQL> shutdown immediate;

SQL> -- start the database according to the modified 20130101. ora static parameter file

SQL> startup pfile = 'd: \ 20120101. ora'

SQL> -- create a new undo tablespace

SQL> create undo tablespace UNDOTBS2 datafile 'data file directory \ undotbs01.dbf' size 2g;

SQL> -- delete the original undo tablespace

SQL> drop tablespace UNDOTBS1 including contents and datafiles;

SQL> -- modify the new tablespace undotbs2 named undotbs1

SQL> alter tablespace UNDOTBS2 rename to UNDOTBS1;

SQL> -- close the database

SQL> shutdown immediate;

SQL> -- restart the database (Problem Solving)

SQL> startup;

The database has now recovered to normal. After 24 hours of operation, no problems have been solved.

PS: it is estimated that the UNDO tablespace physical files copied by the customer during cold Backup will be damaged or the files will be damaged if the database is not completely stopped, the database can be started without the UNDO tablespace. However, if data is modified, the UNDO tablespace fails to be written and the data connection is disconnected.

This article from the "IT people's network life" blog, please be sure to keep this source http://rainyang.blog.51cto.com/469543/1347834