Home > bug > Bug 6767655记录

Bug 6767655记录

February 12th, 2012 Leave a comment Go to comments

BUG记录:
开发库某段时间连续报如下错误:

Errors in file /oracle/admin/johnnydb/bdump/johnnydb_s033_16679.trc:
ORA-07445: 出现异常错误: 核心转储 [kauxs_do_journal()+54] [SIGSEGV] [Address not mapped to object] [0x000000002] [] []
Mon Feb  6 14:36:54 2012
Errors in file /oracle/admin/johnnydb/bdump/johnnydb_s021_20918.trc:
ORA-07445: 出现异常错误: 核心转储 [kauxs_do_journal()+54] [SIGSEGV] [Address not mapped to object] [0x000000002] [] []
Mon Feb  6 14:36:54 2012
Errors in file /oracle/admin/johnnydb/bdump/johnnydb_s024_15936.trc:
ORA-07445: 出现异常错误: 核心转储 [kauxs_do_journal()+54] [SIGSEGV] [Address not mapped to object] [0x000000002] [] []

错误文件里面可以看到是一个分区表的查询SQL引起:

===========================
Current Instatiation Object
===========================
-------------------------------------
INSTANTIATION OBJECT: object=0x109616830
type=""[56] lock=(nil) handle=(nil) body=0x183730508 level=0
flags=SHR/[961] executions=0
 sqltxt(0x16ddaba20)=
SELECT xxoo from one_part_tab where ...;  

如某人所说,开始是百思不得骑姐,google不到,但是metalink上搜到一个BUG:

Bug 6767655: DUMP IN KAUXS_DO_JOURNAL() WHEN DOING CONCURRENT DML AND DDL

已在产品版本中修复    11.2

产品版本    10.2.0.3

平台    46 – Linux x86

PROBLEM:

——–

When having two sessions running concurrently, one doing inserts to a

partitioned table, and the other doing partition maintenance operations,

there is a chance of the DML session core dumping in kauxs_do_journal(), ie.:

ORA-7445: exception encountered: core dump [kauxs_do_journal()+21] [SIGSEGV]

[Address not mapped to object] [0x2] [] []

The problem

DIAGNOSTIC ANALYSIS:

——————–

The problem seems to be a timing issue – or at least there is a timing

component involved. When doing the partition split, an ORA-54 is normally

raised, however by running the split in a loop, it will eventually succeed

(and thus causing problems for the session running the DML).

WORKAROUND:

———–

None found (apart from avoid doing partition maintenance while DML is

performed)

RELATED BUGS:

————-

REPRODUCIBILITY:

—————-

Problem reproduced by customer on 10.2.0.3 (Linux32 and Solaris64).

Reproduced internally on 10.2.0.3 and 11.1.0.6

除了版本有一点点出入外(10.2.0.4),其他情况吻合,后来确认了下,确实有人在那个时间点做分区维护脚本测试,查看dba_objects.LAST_DDL_TIME,时间点吻合,一段时间之后报错没有再出现。

Categories: bug Tags:
  1. No comments yet.
  1. No trackbacks yet.