今天一早检查alert.log,发现以下错误:

Wed Jul 11 23:10:32 2012
Process J001 died, see its trace file
Wed Jul 11 23:10:32 2012
kkjcre1p: unable to spawn jobq slave process 
Wed Jul 11 23:10:32 2012
Errors in file /opt/oracle/admin/ora10g/bdump/ora10g_cjq0_1662996.trc:

trc文件的内容如下

Redo thread mounted by this instance: 1
Oracle process number: 13
Unix process pid: 712918, image: oracle@zjc
*** SERVICE NAME:(SYS$BACKGROUND) 2012-07-09 23:05:04.037
*** SESSION ID:(3373.1) 2011-01-09 11:05:04.037
*** 2012-07-09 2305:04.037
Process J000 is dead (pid=13680712, state=3):
*** 2012-07-09 23:05:10.038

根据提示信息：kkjcre1p: unable to spawn jobq slave process ，可以了解到是系统无法生成job相关的进程而出错的，那么大约有几种可能：

1、参数job_queue_processes（设置过小）

2、参数session和processes（设置的会话数及连接数不能满足业务需求）

3、参数pga_aggregate_target（被耗尽）

4、OS资源被耗尽，如virtual memory

查看SQL> show parameter job_queue_processes
NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
job_queue_processes                  integer     2

Solution:
Make sure you are using SPFILE and then issue the following :

SQL>alter system set JOB_QUEUE_PROCESSES=50 scope=both;