top看到系统的CPU数少于实际数
各位大侠请指教一下,万分感谢。
我遇到一个问题,top中上面CPU信息那部分,只看到1个CPU,并不能看到其他两个(一共三个),ioscan看到cpu状态正常,machinfo也能认到有3个CPU,但是系统的event.log中有CPU0(主CPU)的报警。
那是不是除了CPU0外,另外两个CPU也坏了呢?
请指教
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(9)
呵呵,谢谢各位的回复。当时top中看到的恰恰是CPU0。我换了CPU0,重启了机器,没报警了。
专业啊!
cstm看看
glance里看看呢
实在不行reboot一把,老机器就这样
这动静有点大了 感觉
1. 进到GSP检查ss输出,看看那2个CPU是否deconfigured掉了没有;是的话就换新CPU吧~
2. 第一步如果看到还是configured的话,就先升级一下OnlineDiag到最新版本,HP-UX11.31的系统有一个OnlineDiag的issue,不过你的系统是HP-UX11.23,建议还是先升级一下(需要先升级你的EMS版本)从而排除这个可能性。。
EMS下载地址:
https://h20392.www2.hp.com/porta ... ber=EventMonitoring
Diagnostic and Support Tools下载地址:
https://h20392.www2.hp.com/porta ... ductNumber=B6191AAE
3. 升级完后重启还是出现EMS那个报错并且top还是看不到,按照EMS的建议是手动deconfigure该CPU,这样的话最终还是要换CPU。
PS:1) EMS报的是slot0的CPU,你top看到的应该没有0编号的CPU吧?
2) 你也可以先到EFI收取一下errdump cmc的输出,检查一下该日志的报错时间和信息。
....没有抓取,只是看到一个CPU....
top方便也贴出来呗
呵呵,好的。
machinfo:CPU info:
Number of CPUs = 3
Clock speed = 1300 MHz
CPUID registers
vendor information = "GenuineIntel"
processor serial number = 0x0000000000000000
processor version info = 0x000000001f010504
architecture revision: 0
processor family: 31 Intel(R) Itanium 2 Family Processors
processor model: 1 Intel(R) Itanium 2 processor
processor revision: 5 stepping B1
largest CPUID reg: 4
processor capabilities = 0x0000000000000001
implements long branch: 1
Bus features
implemented = 0xbdf0000060000000
selected = 0x0000000040000000
Bus Lock Signal masked
Cache info:
L1 Instruction: size = 16 KB, associativity = 4
L1 Data: size = 16 KB, associativity = 4
L2 Unified: size = 256 KB, associativity = 8
L3 Unified: size = 3072 KB, associativity = 12
ioscan:
...
processor 0 120 processor CLAIMED PROCESSOR Processor
processor 1 121 processor CLAIMED PROCESSOR Processor
processor 2 122 processor CLAIMED PROCESSOR Processor
...
event.log:
>------------ Event Monitoring Service Event Notification ------------<
Notification Time: Tue Dec 15 10:36:42 2009
rx5670 sent Event Monitor notification information:
/system/events/cpu/cmc/cache_errors is >= 1.
Its current value is SERIOUS(4).
Event data from monitor:
Event Time..........: Tue Dec 15 10:36:42 2009
Severity............: SERIOUS
Monitor.............: cmc_em
Event #.............: 100622
System..............: rx5670
Summary:
Cache Error(s) detected on processor in slot (cpu socket 0 ).
Description of Error:
Due to errors indicated in the prior 100611 EMS Event(s), the monitor
tried to deactivate the processor for current boot session. Since this is
the MONARCH processor on the system, it cannot be deactivated. The monitor
tried to mark it for deconfiguration for the next boot session but was not
successful. Manual deconfiguration of the faulty processor at EFI level is
required.
Probable Cause / Recommended Action:
Contact your HP support representative to check the processor.
Additional Event Data:
System IP Address...: 192.168.0.1
System IP Address...: 132.32.21.16
Event Id............: 0x4b26f63a00000000
Monitor Version.....: B.01.00
Event Class.........: unavailable
Client Configuration File...........:
/var/stm/config/tools/monitor/default_cmc_em.clcfg
Client Configuration File Version...: A.01.00
Qualification criteria met.
Number of events..: 1
Associated OS error log entry id(s):
None
Additional System Data:
System Model Number.............: ia64 hp server rx5670
EMS Version.....................: A.04.00
STM Version.....................: B.40.00
OS Version......................: B.11.23
System Serial Number............:
Latest information on this event:
http://docs.hp.com/hpux/content/hardware/ems/cmc_em.htm
v-v-v-v-v-v-v-v-v-v-v-v-v D E T A I L S v-v-v-v-v-v-v-v-v-v-v-v-v
Processor Information:
======================
Slot Number: cpu socket 0
Physical Location: Unavailable
Cabinet : Not Applicable
Cell Slot : Not Applicable
Slot : Not Applicable
LID: 0x0000000000000000
Dynamic Processor Resilience action threshold met.
Number of parity errors : 10
Received in : 1 Day(s)
>---------- End Event Monitoring Service Event Notification ----------<
top中看不到CPU1和2....
[ 本帖最后由 fengyan 于 2009-12-21 17:07 编辑 ]
好歹把日志贴出来啊