ORA-07445: exception encountered: core dump [ktspsrch_reset()+368] [SIGSEGV]

兜兜

数据库版本：oracle 11.2.0.3.0  64bit
服务器系统：HP-UX  B.11.31 U ia64
内存：64G

Sat Mar 01 03:14:42 2014
Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x9FFFFFFFB5B26030] [PC:0x4000000008952230, $cold_ktspsrch_reset()+368] [flags: 0x0, count: 1]
Errors in file /oradata1/dmpfile/diag/rdbms/pbhadb/PBHADB/trace/PBHADB_ora_26467.trc  (incident=55305):
ORA-07445: exception encountered: core dump [ktspsrch_reset()+368] [SIGSEGV] [ADDR:0x9FFFFFFFB5B26030] [PC:0x4000000008952230] [Address not mapped to object] []
Incident details in: /oradata1/dmpfile/diag/rdbms/pbhadb/PBHADB/incident/incdir_55305/PBHADB_ora_26467_i55305.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Sat Mar 01 03:15:06 2014
Dumping diagnostic data in directory=[cdmp_20140301031506], requested by (instance=1, osid=26467), summary=[incident=55305].
Sat Mar 01 03:15:06 2014
Sweep [inc][55305]: completed
Sweep [inc2][55305]: completed
Sat Mar 01 03:15:34 2014
Error 3113 trapped in 2PC on transaction 7.16.368243. Cleaning up.
Error stack returned to user:
ORA-03113: end-of-file on communication channel
ORA-02063: preceding line from K3_INTERFACE_PRD

兜兜

附件为 trc

兜兜

刘大，麻烦有时间帮忙看一下，我看有点象 Bug 14040124 : ORA-7445 [KTSPSRCH_RESET()+368] 。不敢肯定，如果是这个BUG ，有更好的解决方式么

Maclean Liu(刘相兵

11.2.0.3.0 + HP-UX B.11.31

Dump continued from file: /oradata1/dmpfile/diag/rdbms/pbhadb/PBHADB/trace/PBHADB_ora_26467.trc
ORA-07445: exception encountered: core dump [ktspsrch_reset()+368] [SIGSEGV] [ADDR:0x9FFFFFFFB5B26030] [PC:0x4000000008952230] [Address not mapped to object] []

========= Dump for incident 55305 (ORA 7445 [ktspsrch_reset()+368]) ========
----- Beginning of Customized Incident Dump(s) -----
Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x9FFFFFFFB5B26030] [PC:0x4000000008952230, $cold_ktspsrch_reset()+368] [flags: 0x0, count: 1]



ktcccDeleteCommitCallbacks =>ktspsrchcc_cbk=>cold_ktspsrch_reset

Maclean Liu(刘相兵

ODM FINDING:

ORA-07445: Exception Encountered: Core Dump [ktspsrchcc_cbk()+612] (Doc ID 1488513.1)

PPLIES TO:

Oracle Database - Enterprise Edition - Version 11.2.0.3 to 11.2.0.3 [Release 11.2]
Information in this document applies to any platform.
SYMPTOMS

On 11.2.0.3 the following error occurs.

ORA-07445: exception encountered: core dump [ktspsrchcc_cbk()+612] [SIGSEGV] [ADDR:0xFFFFFFFF76D40838] [PC:0x1028538E4] [Address not mapped to object] []


The call stack in the trace file will contain some or all of the following functions:

... ktspsrchcc_cbk ktcccDeleteCommitCallbacks ktcccdel ktcCommitTxn ktdcmt k2lcom k2send k2serv opiodr ttcpip ..
CAUSE

Bug 14169076 - ORA-7445 [KTSPSRCHCC_CBK()+612]
closed as a duplicate of
Bug 14040124 - ORA-7445 [KTSPSRCH_RESET()+368]

The issue can occur when a local transaction is promoted to a distributed transaction and automatic segment space management is being used.

Bug 14040124 is fixed in 12.1.0.1 (Base Release) and 11.2.0.4 (Server Patch Set).
Note that interim patches for this issue in 11.2 are released under Bug 14313519

Please refer to
Note 14040124.8 - Bug 14040124 - ORA-7445 [ktspsrch_reset] during commit on ASSM segment

SOLUTION

To solve the issue, use any of below alternatives:

o  Upgrade to 12.1

- OR -

o  Apply patch set 11.2.0.4

- OR -

o  Apply interim Patch 14313519> if available for your platform. Note that this patch is not backportable.

- OR -

o  As possible workarounds

- force the application to issue a COMMIT before starting a distributed DML operation

- disable space search cache by setting session event 10019, level 1, if no indexes are involved:

alter session set events '10019 trace name context forever, level 1';

- use MSSM (manual segment space management)


REFERENCES