- 最后登录
- 2013-9-3
- 在线时间
- 5 小时
- 威望
- 0
- 金钱
- 28
- 注册时间
- 2013-6-15
- 阅读权限
- 10
- 帖子
- 12
- 精华
- 0
- 积分
- 0
- UID
- 1128
|
1#
发表于 2013-6-16 14:08:30
|
查看: 3078 |
回复: 2
系统昨天很卡,登录sqlplus要卡个1分钟才能进去, 相关的信息如下:
alert log :
--------------------------------------
Fri Jun 14 19:27:53 2013
Process J000 died, see its trace file
Fri Jun 14 19:27:53 2013
kkjcre1p: unable to spawn jobq slave process
--------------------------------------
trace :
--------------------------------------
*** 2013-05-20 19:02:46.745
Waited for process J006 to be spawned for 120 seconds
*** 2013-05-20 19:07:18.679
*** 2013-05-20 19:08:23.784
Waited for process J006 to initialize for 60 seconds
*** 2013-05-20 19:08:23.784
Dumping diagnostic information for J006:
OS pid = 23767
loadavg : 1.40 2.52 1.81
memory info: free memory = 0.00M
swap info: free = 0.00M alloc = 0.00M total = 0.00M
skgpgpstack: read() for cmd /bin/ps -elf | /bin/egrep 'PID | 23767' | /bin/grep -v grep timed out after 60 seconds
/proc/23767/exe: No such file or directory.
ptrace: No such process.
/u01/app/oracle/product/10.2.0/db_1/dbs/23767: No such file or directory.
No stack.
--------------------------------------
OS info:
--------------------------------------
[oracle@northstar ~]$ cat /etc/redhat-release
Red Hat Enterprise Linux Server release 5.2 (Tikanga)
pmap -d 16859
mapped: 8586728K writeable/private: 45504K shared: 8388632K
Mem: 16777216k total, 16759924k used, 17292k free, 16800k buffers
Swap: 10485752k total, 357188k used, 10128564k free, 15062152k cached
* soft nproc 2047
* hard nproc 16384
* soft nofile 1024
* hard nofile 65536
--------------------------------------
related parameter:
--------------------------------------
sessions integer 170
processes integer 150
--------------------------------------
问题:
--------------------------------------
1. 是不是跟内存cache过大有关?
2. 如果配置大页面内存, 该参数
memlock maximum locked-in-memory address space (KB)
说的是单进程能够lock的最大内存数量 还是所有进程能够lock的内存总量?
--------------------------------------
|
|