Menu

Dipesh Majumdar

Blog and Paintings

Terminating the instance due to error 472

June 27, 2012

It was little puzzling. 

Whenever I was starting the database, the database would start. Remain open for a few seconds and then shut down. 

I saw the alert log and this was the error that was coming:

terminating the instance due to error 472 
Instance terminated by PMON

The alert log showed something like this….
                                                                                                                                                                                                                                                            Thu Jun 21 19:56:59 2012Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_j000_3748.trc  (incident=163651):ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []Incident details in: /u01/app/oracle/diag/rdbms/orcl/ORCL/incident/incdir_163651/ORCL_j000_3748_i163651.trcDoing block recovery for file 3 block 3383Resuming block recovery (PMON) for file 3 block 3383Block recovery from logseq 488, block 61 to scn 9899996Recovery of Online Redo Log: Thread 1 Group 2 Seq 488 Reading mem 0  Mem# 0: /u01/app/oracle/oradata/ORCL/redo02.logBlock recovery completed at rba 488.101.16, scn 0.9900000Thu Jun 21 19:57:01 2012Trace dumping is performing id=[cdmp_20120621195701]Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_j000_3748.trc  (incident=163652):ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []Incident details in: /u01/app/oracle/diag/rdbms/orcl/ORCL/incident/incdir_163652/ORCL_j000_3748_i163652.trcThu Jun 21 19:57:03 2012Sweep [inc][163652]: completedSweep [inc][163651]: completedSweep [inc2][163651]: completedErrors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_j000_3748.trc:ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []Trace dumping is performing id=[cdmp_20120621195705]Trace dumping is performing id=[cdmp_20120621195708]Thu Jun 21 19:57:07 2012Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_j000_3748.trc  (incident=167476):ORA-00603: ORACLE server session terminated by fatal errorORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []Incident details in: /u01/app/oracle/diag/rdbms/orcl/ORCL/incident/incdir_167476/ORCL_j000_3748_i167476.trcErrors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/incident/incdir_167476/ORCL_j000_3748_i167476.trc:ORA-00603: ORACLE server session terminated by fatal errorORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []Thu Jun 21 19:57:10 2012Doing block recovery for file 3 block 3383Resuming block recovery (PMON) for file 3 block 3383Block recovery from logseq 488, block 61 to scn 9899996Recovery of Online Redo Log: Thread 1 Group 2 Seq 488 Reading mem 0  Mem# 0: /u01/app/oracle/oradata/ORCL/redo02.logBlock recovery completed at rba 488.101.16, scn 0.9900000Errors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_pmon_3510.trc  (incident=163489):ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []Incident details in: /u01/app/oracle/diag/rdbms/orcl/ORCL/incident/incdir_163489/ORCL_pmon_3510_i163489.trcErrors in file /u01/app/oracle/diag/rdbms/orcl/ORCL/trace/ORCL_pmon_3510.trc:ORA-00600: internal error code, arguments: [4194], [], [], [], [], [], [], [], [], [], [], []PMON (ospid: 3510): terminating the instance due to error 472Instance terminated by PMON, pid = 3510

I saw all the trace files and other things but didn't get any clue.

Finally, Google came to rescue and this thread helped me:

https://forums.oracle.com/forums/thread.jspa?threadID=2315043

From the above url, I got to know that error was because of undo corruption. However the above url doesn't explain how one can come to know that the problem lies in undo.

I tried to check if my undo was corrupted but couldn’t validate this.

I thought let me change undo tablespace and check for once. This is because I was not getting any clue as to how to solve this error. What I was going to do now was simply an attempt without much logic or sense. Much like a hit and trial attempt.  

So what I did was this –1. Mounted the database. (Didn’t open it)

2. Created a pfile from spfile

3. Edited the pfile (changed UNDOTBS1 to UNDOTBS2).

4. Now created undo tablespace in the mounted state of database with this command –

create UNDO tablespace UNDOTBS2 datafile '/u01/app/oracle/oradata/ORCL/undotbs02.dbf' size 3024m REUSE AUTOEXTEND ON NEXT 4096K MAXSIZE 3024M;

Hang on! The above command won't run in mounted state! The database would open for a few seconds and then terminate. I knew that. So quickly fired the above 'create undo tablespace' command immediately after opening the database. 

5. Shut down the database

6. Restart the database from the pfile just changed.

 Bingo, it worked!!!  I can't believe... blindfolded I have hit the target. Spot on!

Database was now up and running without any problem and there was no further errors in the alert log.

Go Back

Thanks, just saved our development DB with described steps!

Thanks!
We restart our DB after 8h of problems!

extremely happy that this post helped you Francesco! It's an almost 3 year old post and i myself have almost forgotten the steps :-)

Thank you so much. Was struggling to start my DB since last 2 days. Your steps helped it. It did not work during the first attempt. Then the second attempsworked like a magic. Thanks a ton :-)



Comment