oracle虚拟内存命令,11gr2 突然挂掉了，32G的虚拟内存使用96%

11gr2 突然挂掉了！

系统症状：

32G的虚拟内存使用96%， 系统盘硬盘读写100%,  数据文件所在磁盘0%, CPU使用率1%.

在服务器上登录sqlplus都成问题了， 等待好长时间也无法登录进去。 遂紧急强制杀掉oracle smon， pmon 等系统进程， 等待了片刻， sqlplus登录成功， 重启数据库。

类似情况也遇到过一次， 对一个频繁访问的表的一个字段执行了update操作之后， 没有及时提交， 过了没有几分钟， 前台应用基本上全都查询不到数据了， 即使那些根本不需要访问这个表的应用也无法使用了。

查看系统中的现象， 跟这次一样。 提交之后， 过了十来分钟， 系统才恢复正常。

系统down机时间应该在15点左右， 查看alert日志， 截取如下信息片段：

Fri Dec 14 10:56:00 2012

Completed checkpoint up to RBA [0x3917.2.10], SCN: 6924009416

Fri Dec 14 10:57:18 2012

opiodr aborting process unknown ospid (41484394) as a result of ORA-609

Fri Dec 14 10:57:18 2012

opiodr aborting process unknown ospid (50463196) as a result of ORA-609

Fri Dec 14 10:57:18 2012

opiodr aborting process unknown ospid (43384992) as a result of ORA-609

Fri Dec 14 14:00:40 2012

Beginning log switch checkpoint up to RBA [0x391a.2.10], SCN: 6925017630

Thread 1 advanced to log sequence 14618 (LGWR switch)

Current log# 1 seq# 14618 mem# 0: /data/oradata/o10g/redo01a.log

Current log# 1 seq# 14618 mem# 1: /data_duplex/oradata/o10g/redo01c.log

Fri Dec 14 14:00:47 2012

Archived Log entry 15137 added for thread 1 sequence 14617 ID 0x9edfe58 dest 1:

Archived Log entry 15138 added for thread 1 sequence 14617 ID 0x9edfe58 dest 2:

Fri Dec 14 14:05:45 2012

Completed checkpoint up to RBA [0x391a.2.10], SCN: 6925017630

Fri Dec 14 14:55:45 2012

Time drift detected. Please check VKTM trace file for more details.

Fri Dec 14 14:57:07 2012

opiodr aborting process unknown ospid (12976608) as a result of ORA-609

Fri Dec 14 14:57:21 2012

opiodr aborting process unknown ospid (18481186) as a result of ORA-609

Fri Dec 14 14:57:30 2012

opiodr aborting process unknown ospid (8847846) as a result of ORA-609

Fri Dec 14 14:57:30 2012

opiodr aborting process unknown ospid (65994930) as a result of ORA-609

Fri Dec 14 14:57:27 2012

opiodr aborting process unknown ospid (45089018) as a result of ORA-609

ORA-00020: maximum number of processes (800) exceeded

ORA-20 errors will not be written to the alert log for

Fri Dec 14 15:12:07 2012

Fri Dec 14 15:10:13 2012

ORA-00020: maximum number of processes (800) exceeded

opiodr aborting process unknown ospid (24707344) as a result of ORA-609 ORA-20 errors will not be written to the alert log for

Fri Dec 14 15:13:27 2012

Fri Dec 14 15:13:27 2012

opiodr aborting process unknown ospid (2162960) as a result of ORA-609

opiodr aborting process unknown ospid (15335484) as a result of ORA-609

the next minute. Please look at trace files to see all

Fri Dec 14 15:12:31 2012

ORA-20 errors will not be written to the alert log for

ORA-20 errors will not be written to the alert log forFri Dec 14 15:12:07 2012

Fri Dec 14 15:13:21 2012

ORA-00020: maximum number of processes (800) exceeded

opiodr aborting process unknown ospid (48890100) as a result of ORA-609

帮忙大概判断一下， 谢谢！