1748194 會員
3486 線上
108759 解決方案
發表新文章

SCSI EVENT

 
watermelonyu
教授

SCSI EVENT

There are 2 nodes in TruCluster 5.1B. When we got the EVM ALERT as the attachment from node 2, these 2 nodes both hung. What's going on?
4則回覆 4
watermelonyu
教授

SCSI EVENT

check your scsi termincation or cable, it's very possible you have cable fault. maybe just cable excoriate

watermelonyu
教授

SCSI EVENT

剛換過 SCSI 卡, 狀況還是一樣
watermelonyu
教授

SCSI EVENT

scsi接線換一條吧, 再者檢查你的scsi terminal有沒有壞掉.
watermelonyu
教授

SCSI EVENT

Hi Eric,



> Sequence number of error: 1485832687

> Time of error entry: 10-Dec-2004 17:13:51

> Host name: hostname

>

> SCSI CAM ERROR PACKET

> SCSI device class: DEC SIM

> Bus Number: 5

> Target number: 4

> Lun Number: 0

>

> Name of routine that logged the event: ss_perform_timeout

> Event information: timeout on disconnected request



The disk 5/4/0 (B/T/L) had failed to access after it encountered a ss_perform_timeout. Please using the command "wsea" to analyze this HW root cause or calling HP engineer for further analysis.



PS: 1) Please check it has the other disks at the same bus 5 via "hwmgr". If no, it needs to check SCSI card, cable, terminate....again. Otherwise,

only this problem disk 5/4/0 needs to investigate.

2) If this problem disk 5/4/0 is accessed as "/",'/usr", "/var" or these busy common processes runon cluster members. It should be caused both members hung as you said.



Best regards,

Richard.