Tales From A Lazy Fat DBA

Recently while opening a TEST RAC database we found the database down and was not available, we tried to start the database but received communication errors and left us clueless. Due to some urgent POC activity scheduled on the database we started investigating the root cause right away for error.
ORA-03113: end-of-file on communication channel

[oracle@testdbdixit ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.4.0 Production on Thu Feb 26 01:04:23 2015
Copyright (c) 1982, 2013, Oracle. All rights reserved.

Connected to an idle instance.
SQL> startup
ORA-03113: end-of-file on communication channel

We tried to open the database in MOUNT mode and it reached to phase easily without any error.

SQL> STARTUP MOUNT;
ORACLE instance started.

Total System Global Area 1.6034E+10 bytes
Fixed Size 2269072 bytes
Variable Size 2449473648 bytes
Database Buffers 1.3556E+10 bytes
Redo Buffers 26480640 bytes
Database mounted.

But the recovery of any kind got failed because of the database running in No Archivelog Mode & being a test instance there is also not any RMAN backups configured as well … #TotalDisaster 😦 😦

SQL> ALTER DATABASE RECOVER DATABASE UNTIL CANCEL;
ALTER DATABASE RECOVER DATABASE UNTIL CANCEL
*
ERROR at line 1:
ORA-00279: change 7311130 generated at 02/25/2015 22:00:18 needed for thread 2
ORA-00289: suggestion : +FRA
ORA-15173: entry ‘ARCHIVELOG’ does not exist in directory ‘DIXITDB’
ORA-00280: change 7311130 for thread 2 is in sequence #207

SQL> archive log list
Database log mode No Archive Mode
Automatic archival Disabled
Archive destination USE_DB_RECOVERY_FILE_DEST
Oldest online log sequence 233
Current log sequence 234

When tried to start the cancel based incomplete recovery, it went successful but received few inconsistency errors for system datafile if we try to open the database in RESETLOGS mode.

SQL> ALTER DATABASE RECOVER CANCEL;
ALTER DATABASE RECOVER CANCEL
*
ERROR at line 1:
ORA-01547: warning: RECOVER succeeded but OPEN RESETLOGS would get error below
ORA-01194: file 1 needs more recovery to be consistent
ORA-01110: data file 1: ‘+DATA/DIXITDB/datafile/system.256.871197447’

As expected RESETLOGS mode failed too.

SQL> ALTER DATABASE OPEN RESETLOGS;
ALTER DATABASE OPEN RESETLOGS
*
ERROR at line 1:
ORA-01194: file 1 needs more recovery to be consistent
ORA-01110: data file 1: ‘+DATA/DIXITDB/datafile/system.256.871197447’

Resolution:
====================
*Note: Underscore (hidden/undocumented) parameters should only be used after consent with Oracle Support and should always tried and tested in sandbox environments before applying them in prod environment.

There is a hidden parameter _ALLOW_RESETLOGS_CORRUPTION=TRUE which will allow us to open database even though it’s not properly recovered.
ALTER SYSTEM SET “_allow_resetlogs_corruption”= TRUE SCOPE = SPFILE;
Tip: Also change the undo_management to “Manual”

After the two changes in the spfile you can open the database with:

sqlplus “/ as sysdba”
startup force

Note: Well there is no 100% guarantee that setting _ALLOW_RESETLOGS_CORRUPTION=TRUE will open the database. However, once the database is opened, then we must immediately rebuild the database. Database rebuild means doing the following, namely: (1) perform a full-database export, (2) create a brand new and separate database, and finally (3) import the recent export dump. This option can be tedious and time consuming, but once we successfully open the new database, then we expect minimal or perhaps no data loss at all. Before you try this option, ensure that you have a good and valid backup of the current database.

Previous Settings:
SQL> show parameter undo

NAME TYPE VALUE
———————————— ———– ——————————
undo_management string AUTO
undo_retention integer 900
undo_tablespace string UNDOTBS1

SQL> alter system set undo_management=manual scope=spfile;
System altered.

SQL> ALTER SYSTEM SET “_allow_resetlogs_corruption”= TRUE SCOPE = SPFILE;
System altered.

SQL> shut immediate;

And after setting all the requisite parameters to defined values, we finally saw that ‘Database Opened’ message on the SQL prompt … 🙂 🙂

SQL> startup force;
ORACLE instance started.

Total System Global Area 1.6034E+10 bytes
Fixed Size 2269072 bytes
Variable Size 2449473648 bytes
Database Buffers 1.3556E+10 bytes
Redo Buffers 26480640 bytes
Database mounted.
Database opened.

SQL> ALTER DATABASE OPEN RESETLOGS;
Database altered.

SQL> archive log list
Database log mode No Archive Mode
Automatic archival Disabled
Archive destination USE_DB_RECOVERY_FILE_DEST
Oldest online log sequence 1
Current log sequence 2

SQL> alter system switch logfile;
System altered.

And a new incarnation of the database happened after opening database in RESETLOGS mode.

SQL> select INCARNATION#,RESETLOGS_TIME,STATUS, RESETLOGS_ID from v$database_incarnation;

INCARNATION# RESETLOGS STATUS RESETLOGS_ID
———— ——— ——- ————
1 24-AUG-13 PARENT 824297850
2 09-FEB-15 PARENT 871197521
3 26-FEB-15 CURRENT 872646322

In Short:
=====================
1) Set _ALLOW_RESETLOGS_CORRUPTION=TRUE in init.ora file.
2) Startup Mount
3) Recover database
4) Alter database open resetlogs.
5) reset undo_management to “manual” in init.ora file.
6) startup database
7) Create new undo tablespace (There is a strict need of changing the UNDO_MANAGEMENT to AUTO in Prod databases)
change undo_management to “AUTO” and undo_tablespace to “NewTablespace”
9) Bounce database.

Hope That Helps!
Prashant Dixit