This post is more than 5 years old
6 Posts
0
44091
January 9th, 2013 09:00
Replacing failed RAID 1 drive - PE1900
Hello all,
I have a PowerEdge 1900 server with two mirrored drives. One of the drives went bad, and OpenManage sees it as Critical/Failed. I have a replacement drive ready to go, I just want to make sure that I'm doing things right before I start.
In order to swap this drive, I'm going to have to turn off the machine as it's in a drive bay that needs to be accessed from within. If it were hotswappable, I wouldn't have my first concern which is: If I just turn this machine off, physically replace the drive (bad drive out, good drive in same slot) and reboot, will I be able to get back to OpenManage to configure it to rebuild?
Are there things I should do before powering down?
Should I be going into the Ctrl-R controller setup on boot instead and doing everything there?
My big fear is that it won't boot, or that the new blank drive will overwrite the good installed one erasing everything or I'll be eaten by gremlins or something.
Thanks in advance,
John
Daniel My
12 Elder
•
6.2K Posts
1
January 9th, 2013 10:00
Hello John
Yes, that will work.
The only time you want to be concerned with swapping out a drive with the server offline is if the drive was in an online state. Our SAS controllers are better at checking the meta data on all of the drives to detect the state of the array, so the issue is mainly only with our old SCSI controllers. If you swap out an online drive while the server is off then the controller will not recognize the drive was removed, and when the server is powered back on it will treat the drive as a normal online drive and start copying the blank data across the array. This will corrupt the array and make it unusable. This isn't really a problem with our newer controllers, but I would still make sure to offline any drives before replacing them with the server powered down just to be safe.
Since this drive is already in an offline state you shouldn't have to worry about this. Just swap out the drive and then set it as a hot spare in OMSA when you turn the server back on. It should then be pulled into the array and initiate a rebuild.
Thanks
jlatour
6 Posts
0
January 9th, 2013 11:00
Awesome. Going to take care of it shortly. Wish me luck.
I'll try to report back with the results, good or bad, for future forum searchers.
John
jlatour
6 Posts
0
January 10th, 2013 07:00
Worked as planned. Thanks again.