A bug found

Got an error from alert log:

Tue Mar  8 04:54:18 2011
Errors in file /global/oracle1/centDB/admin/centDB/udump/centdb_ora_8749.trc:
ORA-07445: exception encountered: core dump [_memcmp()+88] [SIGSEGV] [Address not mapped to object] [0x000000010] [] []

 

SQL> SELECT NAME, VALUE
  2  FROM V$SYSSTAT
  3  WHERE NAME = 'redo buffer allocation retries';

NAME                                                                  VALUE
---------------------------------------------------------------- ----------
redo buffer allocation retries                                       387582

SQL> SELECT a.name,b.value
  2  FROM v$statname a,v$sysstat b
  3  WHERE a.statistic# = b.statistic#
AND a.name like '%redo%';  4 

NAME                                                                  VALUE
---------------------------------------------------------------- ----------
redo synch writes                                                  42621021
redo synch time                                                    42146939
redo blocks read for recovery                                             0
redo entries                                                      229665272
redo size                                                        3.7509E+11
redo buffer allocation retries                                       387582
redo wastage                                                     5160641384
redo writer latching time                                              1297
redo writes                                                        27577242
redo blocks written                                               762220497
redo write time                                                    50104443

NAME                                                                  VALUE
---------------------------------------------------------------- ----------
redo log space requests                                                8977
redo log space wait time                                             273809
redo log switch interrupts                                                0
redo ordering marks                                                19457235
redo subscn max counts                                                    0

16 rows selected.

Top 5 Timed Events                                         Avg %Total
~~~~~~~~~~~~~~~~~~                                        wait   Call
Event                                 Waits    Time (s)   (ms)   Time Wait Class
------------------------------ ------------ ----------- ------ ------ ----------
log file parallel write               6,467         109     17   73.5 System I/O
log file sync                         6,411          75     12   50.5     Commit
db file parallel write                1,301          70     53   47.0 System I/O
CPU time                                             49          33.3
log buffer space                         54          40    750   27.4 Configurat

 

But the database is not busy during the timeframe:

              Snap Id      Snap Time      Sessions Curs/Sess
            --------- ------------------- -------- ---------
Begin Snap:     38796 08-Mar-11 04:00:06        38       1.2
  End Snap:     38797 08-Mar-11 05:00:08        38       1.3
   Elapsed:               60.02 (mins)
   DB Time:                2.47 (mins)

 

So I doubt it is a bug:

SQL> select * from v$version;

BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bi
PL/SQL Release 10.2.0.1.0 - Production
CORE    10.2.0.1.0      Production
TNS for Solaris: Version 10.2.0.1.0 - Production
NLSRTL Version 10.2.0.1.0 - Production

Searched it from MOS and found Bug 4591936

Applies to:

Oracle Server - Enterprise Edition - Version: 10.1.0.2 to 10.2.0.3
This problem can occur on any platform.

Symptoms

Trace files are getting generated with following errors logged in trace files:

*** 2006-06-13 22:24:29.826
*** SERVICE NAME:() 2006-06-13 22:24:29.825
*** SESSION ID:(166.1) 2006-06-13 22:24:29.825
Maximum redo generation record size = 132096 bytes
Maximum redo generation change vector size = 98708 bytes
*** 2006-06-13 22:24:57.649
Media recovery not enabled or manual archival only 0x10000
Media recovery not enabled or manual archival only 0x10000
Media recovery not enabled or manual archival only 0x10000
Media recovery not enabled or manual archival only 0x10000

Cause

Bug 4591936
Abstract: KCCSGA_UPDATE_CKPT,KCCDEBUG_LEVEL,MEDIA RECOVERY MESSAGES ARE LOGGED AT STARTUP
Fixed In Ver: 11.0

Solution

These messages were added for debugging purpose and can be ignored.
Delete these files manually or use some job to clean up diskspace.

A one-off patch is available on top of 10.2.0.3 for following platforms
- HP-UX Itanium
- Solaris 64 Bit

Bug is fixed in 10.2.0.4, 11.1.0.6 and higher versions.

To download:

1. Go to metalink.oracle.com
2. On top left hand side under Quick Find from the drop down list choose "Patch Number" and provide the patch number as 4591936 in the box below it. Click go.
3. In the following page, choose your  platform and download the patch. ( This will list all the plaforms for which patch is available )

Please follow the readme.txt for instruction on how to apply the patch.

Please note that these one-off patches are not regression tested and hence you are advised to apply the same on a test server before applying it on a production box.

References

BUG:4591936 - KCCSGA_UPDATE_CKPT,KCCDEBUG_LEVEL,MEDIA RECOVERY MESSAGES ARE LOGGED AT STARTUP

---