Wednesday, March 16, 2016

ORA-07445 kewa_dump_time_diff() , Database does not startup after shutdown abort

ORA-07445 kewa_dump_time_diff() , Database does not startup after shutdown abort

The database will not start and ORA-07445: exception encountered: core dump [kewa_dump_time_diff()+108] errors are reported on database startup.

The alert.log will show errors similar to:

ORA-07445: exception encountered: core dump [kewa_dump_time_diff()+108] [SIGSEGV] [ADDR:0x4CD0]

[PC:0x102A8184C] [Address not mapped to object] []

Incident details in:

/RD0/app/oracle/admin/diag/rdbms/smurfuat/SMURFUAT/incident/incdir_33609/SMURFUAT_ora_26563_i33609.trc

Sweep Incident[33609]: completed

Mon Jun 22 10:57:52 2009

ORA-1092 : opitsk aborting process

Mon Jun 22 10:58:29 2009

License high water mark = 32

WARNING: process USER (ospid: 637) was unable to attach SMR.

Errors in file ...:

ORA-27300: OS system dependent operation:open failed with status: 2

ORA-27301: OS failure message: No such file or directory

ORA-27302: failure occurred at: sskgmsmr_7

USER (ospid: 637): terminating the instance

Termination issued to instance processes. Waiting for the processes to exit

Mon Jun 22 10:58:41 2009

Instance termination failed to kill one or more processes

Mon Jun 22 10:59:06 2009

Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x4CD0] [PC:0x102A8184C,

kewa_dump_time_diff()+108]

Errors in file ...

(incident=33610):

ORA-07445: exception encountered: core dump [kewa_dump_time_diff()+108] [SIGSEGV] [ADDR:0x4CD0]

[PC:0x102A8184C] [Address not mapped to object] []

Incident details in:

...

Mon Jun 22 11:05:25 2009

ORA-1092 : opitsk aborting process

Mon Jun 22 11:08:57 2009

ORA-1092 : opitsk aborting process

Mon Jun 22 11:14:22 2009

Cause:

The cause of this error has been identified as an issue with the shared memory or semaphores left around after the last shutdown.

There are 2 bugs that have been opened in relation to this error:

Unpublished Bug 8503420 which was closed as not a bug as development resolved this erros  by having the customer remove the left around shared memory and semaphores

Unpublished Bug 7157404 which is fixed in the 11.2.0.1 database version.  The bug causes this error to be reported if the pointer to the SGA is null due to shared memory segments not being cleaned up during the previous shutdown.

Both bugs show the ora-7445 [kewa_dump_time_diff] error. The database will not start which is the primary issue in the bugs.


Solution:

To implement the solution, please execute the following steps:

Check to verify there are no shared memory or semaphores owned by Oracle with the following

command:

$ ipcs

If there are any shared memory or semaphores owned by oracle remove with the following


command:

To delete shared memory: % ipcrm -m <shared memory id>

To delete semaphores: % ipcrm -s <semaphore id>


If you have multiple instances running on the same server then you can use the following note to determine which shared memory or semaphores will need to be removed. By using the information within the note, you will not need to take all instances down to remove the 'bad' shared memory or semaphores:

DETERMINING WHICH INSTANCE OWNS WHICH SHARED MEMORY & SEMAPHORE SEGMENTS  Note 68281.1

Reference Metalink Doc ID 847786.1

No comments:

Post a Comment