刀箱UIS8000中配置8台刀片服务器安装 Centos6.8 做群集 ,当出现高CPU\内存负载时,8台刀片会有随机性出现1~2个刀片服务器宕机,宕机刀片健康灯橙色报警,重启后刀片服务器可以正常开机运行,无报警,但故障会不定期反复出现。
出现故障时,刀片健康灯报警。
IML log可见以下报错内容
Server Blade Enclosure Inadequate Power To Power On: Not Enough Power (Enclosure Serial Number xxxxxxxxxx Slot 8)
Maintenance note: CPU(s) operating at reduced performance level due to an enclosure power event.
OA SysLog可见刀片健康状态报警
Mar 14 17:39:35 OA: Blade 8 is reporting failed health status.
Mar 14 17:39:35 OA: Blade in bay #8 status changed from OK to Failed
Mar 14 17:41:40 OA: Blade 1 is reporting failed health status.
Mar 14 17:41:40 OA: Blade in bay #1 status changed from OK to Failed
查看刀箱OA中电源设置,Dynamic Power是开启状态,由于大部分时间刀片工作在低负载状态,刀箱自动降低对刀片的供电输出,刀箱对刀片供电提高需要响应时间,当系统瞬间出现高负载时,刀片硬件从刀箱获得供电不够导致硬件异常。
查看刀箱电源配置方式:
>SHOW POWER
Power Mode: Redundant
Dynamic Power: Enabled
Set Power Limit: Not Set
OA中关闭Dynamic Power功能,观察使用故障无复现。
当出现同刀箱内多台刀片供电相关问题时,可以优先考虑刀箱电源设置,是否有限制,或者动态电源管理。
该案例暂时没有网友评论
✖
案例意见反馈
亲~登录后才可以操作哦!
确定你的邮箱还未认证,请认证邮箱或绑定手机后进行当前操作