1 Rookie
•
27 Posts
0
32268
January 3rd, 2022 10:00
Correct way to replace drive in RAID
I have an R710/H700 controller and a single RAID 5 running on 4 drives with ESXi 6.5 as the OS. I've noticed one of the RAID5 drives is blinking orange.
What is the correct method to replace a failing drive on the PERC H700? Do I just pull the bad drive, wait for 30 seconds, then put a new one in? I've done this on other PowerEdge systems and it was just a hot-swap.
0 events found
No Events found!
DELL-Shine K
6 Operator
•
3K Posts
0
January 4th, 2022 05:00
Sorry, I did not notice that your drive is in failed state. If drive is in failed state then you can replace the drive without making offline.
Steps will be
DELL-Young E
Moderator
•
5.4K Posts
•
37 Points
0
August 20th, 2024 05:18
Hello,
"Oddly enough, I pulled the drive and just reinserted it and everything was fine. It rebuilt the array using the existing drive and there are no more reports of a failed drive. " That indicates the drive may have failed because of bad connection also considering the age of the system anything could happen.
If you pull the drive and reinsert it either will start using hot spare
or if no hot spare will try to rebuild the same drive after it goes in ready state.
If anyone wants to find out more the best way to see what exactly happened was to have your TTY logs analyzed.
Respectfully,
hpcTech
1 Rookie
•
27 Posts
0
January 3rd, 2022 10:00
I read that it should be hot-swappable since it's on the SAS backplane. However, my other system didn't report a failing drive and when the machine was powered off, a second drive died - killing the RAID 5 array. I ended up replacing all 4 drives with larger ones and recreating the array.
DELL-Young E
Moderator
•
5.4K Posts
•
37 Points
1
January 3rd, 2022 20:00
Hi hpcTech, happy 2022 to you!
I am glad you got it sorted eventually so I'm only leaving this as future reference for others to see.
https://dell.to/3EWCH1n
Have a good one!
hpcTech
1 Rookie
•
27 Posts
0
January 4th, 2022 04:00
I do not have it sorted. How long do I pull the drive for? Is there a specific time?
DELL-Shine K
6 Operator
•
3K Posts
0
January 4th, 2022 05:00
Recommended way is to make the drive offline (You can user PERC BIOS or OMSA for this) first then replace the drive. If the rebuild not started automatically after replacing the drive you can assign the drive as dedicated hot spare for the virtual disk. Please ensure new disk is same type and family of existing drive.
It is good practice to backup your data before carrying out any maintenance tasks.
hpcTech
1 Rookie
•
27 Posts
0
January 4th, 2022 05:00
Is there a way you could run through the tasks? I'm not familiar marking a drive offline in the PERC setup. I thought I could hot swap the drives? I specifically posted this was ESXi 6.5 to let people know OMSA is not an option.
IE:
1. Reboot the server and hit CTL+R to get into PERC setup
2. Mark the physical drive offline by hitting F2 on the failing drive then power down
3. Remove the drive, replace with another, boot system
4. Get back into PERC setup...
hpcTech
1 Rookie
•
27 Posts
1
January 9th, 2022 03:00
Oddly enough, I pulled the drive and just reinserted it and everything was fine. It rebuilt the array using the existing drive and there are no more reports of a failed drive.
Thanks for your help.
JustinKPFlow
1 Message
•
2 Points
0
August 19th, 2024 23:32
@hpcTech Yea, just did that on mine, and started rebuilding right away... Even after two reboots, it did not kick off by itself, i guess removing the drive and plugging it back in jump starts the "Hot Swap" flag!
Thanks Much!