环境
   oracle 11.2
   aix 6.1
两个数据库之间使用dblink+impdp进行数据迁移，迁移过程中出现如下错误。
出现这个错误的原因是，dblink+数据泵进行数据加载的过程中会把源库的scn同步到目标库上，造成目标库scn增长过快。

-----数据库告警日志

Rejected the attempt to advance SCN over limit by 21 hours worth to 0x0cae.4d0923b9, by distributed transaction remote

logon, remote DB: RKKDB.
Client info : DB logon user ORA_RECO_070361, machine ora2a, program oracle@ora2a (RECO), and OS user oracle
Errors in file /u01/oracle/app/diag/rdbms/gkhxk/gkhxk2/trace/gkhxk2_reco_9371650.trc:
ORA-19706: invalid SCN
Rejected the attempt to advance SCN over limit by 21 hours worth to 0x0cae.4d092422, by distributed transaction remote

logon, remote DB: RKKDB.
Client info : DB logon user ORA_RECO_070361, machine ora2a, program oracle@ora2a (RECO), and OS user oracle
Errors in file /u01/oracle/app/diag/rdbms/gkhxk/gkhxk2/trace/gkhxk2_reco_9371650.trc:
ORA-19706: invalid SCN
Rejected the attempt to advance SCN over limit by 21 hours worth to 0x0cae.4d09248a, by distributed transaction remote

logon, remote DB: RKKDB.
Client info : DB logon user ORA_RECO_070361, machine ora2a, program oracle@ora2a (RECO), and OS user oracle
Errors in file /u01/oracle/app/diag/rdbms/gkhxk/gkhxk2/trace/gkhxk2_reco_9371650.trc:
ORA-19706: invalid SCN
Rejected the attempt to advance SCN over limit by 21 hours worth to 0x0cae.4d0924d7, by distributed transaction remote

logon, remote DB: RKKDB.
Client info : DB logon user ORA_RECO_070361, machine ora2a, program oracle@ora2a (RECO), and OS user oracle
Errors in file /u01/oracle/app/diag/rdbms/gkhxk/gkhxk2/trace/gkhxk2_reco_9371650.trc:

----跟踪文件日志

#########################trace:#####################

Trace file /u01/oracle/app/diag/rdbms/gkhxk/gkhxk2/trace/gkhxk2_reco_9371650.trc
Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP,
Data Mining and Real Application Testing options
ORACLE_HOME = /u01/oracle/app/oracle/11.2.4/db_1
System name:    AIX
Node name:      ora2a
Release:        1
Version:        6
Machine:        00F9205C4C00
Instance name: gkhxk2
Redo thread mounted by this instance: 2
Oracle process number: 59
Unix process pid: 9371650, image: oracle@ora2a (RECO)
*** 2014-06-23 15:15:46.934
*** SESSION ID:(4190.1) 2014-06-23 15:15:46.934
*** CLIENT ID:() 2014-06-23 15:15:46.934
*** SERVICE NAME:(SYS$BACKGROUND) 2014-06-23 15:15:46.934
*** MODULE NAME:() 2014-06-23 15:15:46.934
*** ACTION NAME:() 2014-06-23 15:15:46.934

ERROR, tran=53.32.68, session#=1, ose=0:
ORA-19706: invalid SCN

*** 2014-06-23 15:15:47.157
ERROR, tran=33.15.204, session#=1, ose=0:
ORA-19706: invalid SCN

*** 2014-06-23 15:15:47.345
ERROR, tran=92.21.280, session#=1, ose=0:
ORA-19706: invalid SCN

*** 2014-06-23 15:15:47.502
ERROR, tran=10.0.4918, session#=1, ose=0:
ORA-19706: invalid SCN

*** 2014-06-23 15:15:47.628
ERROR, tran=27.33.154, session#=1, ose=0:
ORA-19706: invalid SCN

-----------问题解决：

alter system set _external_scn_rejection_threshold_hours=2 scope=spfile
从新启动数据库后数据加载正常。