Today, we share a maintenance case:
Machine Model: SA5212M4, array is LSI card, the initial level of the array card, hard disk has a small problem will not alarm. This machine has a rear drive, which belongs to the custom machine.
Today, the operation of e-Mail informed that there is a wave server hard disk failure. But our on-site patrol did not find a problem, no error, status lights are normal.
At that time I think of the hard drive can have bad, read and write has a large delay, the system to determine the hard disk failure. Send content according to email:
a computer room host 172.27.12.58 disk/dev/sdl failure. Now there is a doubt, the scene no alarm, do not know which block hard drive bad, this nasty. Later thought can be through the hard drive serial number to find out which piece of bad hard drive. Through the command can smartctl-a/dev/sdl/, find the hard disk serial number, and then according to the location of the hard disk, SDA corresponds to the first hard disk, and so on, locate the hard disk, operation provides the hard drive serial number, check whether consistent.
Here are the steps I summarized:
Factory personnel to the scene----"e-mail notification operation and maintenance (emergency call)----" The scene of the machine, and the display view status, self-test error in the hard drive does not show please contact operation to view the hard drive serial number (need to power on and off, and email us serial number)-----" Replace the hard drive serial number after checking.
The above is personal experience, pure personal notes, not good spray!
This article is from the "Pcjazz" blog, make sure to keep this source http://520527.blog.51cto.com/510527/1880697
An example of a wave SA5212M4 server hard drive replacement.