急！一个快照的错误，请各位帮忙啊！

老技术员

在刷新的快照时，出现以下错误
SQL>; EXECUTE DBMS_SNAPSHOT.REFRESH('QB_SJR_BW','C');
BEGIN DBMS_SNAPSHOT.REFRESH('QB_SJR_BW','C'); END;

*
ERROR 位于第 1 行:
ORA-02055: 分布式更新操作失效；要求重算
ORA-06512: 在"SYS.DBMS_SNAPSHOT", line 604
ORA-06512: 在"SYS.DBMS_SNAPSHOT", line 661
ORA-06512: 在"SYS.DBMS_SNAPSHOT", line 641
ORA-06512: 在line 1

是什么原因 请各位帮忙！！！

CCBZZP

Cause: A failure during a distributed update operation may not have rolled back all
effects of the operation. Since some sites may be inconsistent with others, the
transaction must roll back to a savepoint or roll back entirely.

Action: Roll back to a savepoint or roll back the entire transaction and resubmit the
rolled-back statements.

老技术员

谢谢！
我想强制使之一致，（以前的操作及数据不是很重要），应该怎么做？  
（不好意思，能否用汉语回帖吗！？   ）

allinwu

If a database at a master site has an SCN that is big enough to have a non-zero
wrap number, then doing a snapshot refresh generates an ORA-07445 error and a
core dump on the master site, and an ORA-02055 on the snapshot site.  This
happens regardless of the mechanism used to do the refresh, whether with the
DBMS_SNAPSHOT.REFRESH() procedure, or with automatic snapshot updating (via
SNPn processes and the INIT.ORA parameters JOB_QUEUE_PROCESSES and
JOB_QUEUE_INTERVAL).  It also happens regardless of the method used, whether
fast, complete, or force.

Workaround:
None.  You can drop and re-create the snapshot; but this is not really a
valid workaround, because then there is no point to having the snapshot.

How to determine rediscovery:
Look at the snapshot site where the REFRESH was performed for errors such as
the following:

SQL>; exec dbms_snapshot.refresh( 'xyz' , 'f' )
ORA-02055: distributed update operation failed; rollback required
ORA-06512: at "SYS.DBMS_SNAPSHOT", line 604
ORA-06512: at "SYS.DBMS_SNAPSHOT", line 661
ORA-06512: at "SYS.DBMS_SNAPSHOT", line 641
ORA-06512: at line 1

If automatic refresh is being used, then look in the alert log on the snapshot
site for errors such as the following:

Fri Sep 24 19:54:55 1999
Errors in file /abc/oracle/rdbms/log/xx_snp0_16952.trc:
ORA-12012: error on auto execute of job 1
ORA-02055: distributed update operation failed; rollback required
ORA-06512: at "SYS.DBMS_SNAPSHOT", line 604
ORA-06512: at "SYS.DBMS_SNAPSHOT", line 661
ORA-06512: at "SYS.DBMS_IREFRESH", line 577
ORA-06512: at "SYS.DBMS_REFRESH", line 211
ORA-06512: at line 1

In either case, you should also look in the alert log on the master site for an
ORA-7445 such as the following:

Wed Sep 22 15:46:24 1999
Errors in file /abc/oracle/rdbms/log/xx_ora_12072.trc:
ORA-07445: exception encountered: core dump [kcmstn()+304] [SIGSEGV]
   [Address not mapped to object] [0] [] []

Look in the trace file on the master site for a stack trace like this:

  ttcpip =>; opiodr =>; kporpc =>; kkxrpc =>; prient =>; prient2 =>;
pricbr =>;
  pricar =>; pfrrun =>; pefcal =>; pefccal =>; psdextp =>; rpiswu2 =&
gt; psdexsp =>;
  peftrusted =>; spefmccallstd =>; spefcmpa =>; kkzibvl =>; kcmstn =>;

  sigacthandler =>; ssexhd =>; ksedmp

Check that the database SCN has a non-zero wrap number with this query:

  select checkpoint_change# / 4294967296 as wrap from v$database ;

If the result is one or more, then you have a non-zero wrap number and you hit
this bug.

Initial broken version:
8.1.5  The bug does not occur in any version of 8.0.



fix:

Fixed in Oracle 8.1.7
Fix will be included in patchsets 8.1.5.2 and 8.1.6.1. Several one-off patches
provided, refer to the bug for additional information.

老技术员

非常感谢！！！

Super_Aladdin

我也遇到同样的问题，大家帮忙解决一下吧

Super_Aladdin

重建快照日志似乎也解决不了