Replacing failed disk with another server disk on the fly with no downtime ?

[Server-Support]

Hi All,

Is it possible to replace the failed disk in one of my server with another disk (which contains data from not so important server) on the fly without reboot or down time ?

[Michael A. McKenney · Accepted solution]

Server type? Is it RAID? Is it hot swap chassis or internal drive?

[Server-Support]

This is HP DL 380 G3 with 2x 76 GB RAID 1 HDD.

can I safely assume that inserting spare hard disk with the existing old data in it can be done to recalculating the RAID set on the fly safely ?

[Michael A. McKenney]

You already have RAID 1 in place. You want to boot with the drive with data on it. You would need to mount that drive first. Boot it so the OS on it loads. Then add it to RAID 1. If you add it with a drive already in it, you would probably lose the data on it.

If the RAID 1 array is failed. You need to break it. Then you might be able to add the second drive without it erasing and rebuilding the array.

I would not do it on the fly. Unless you want to lose data.

[Server-Support]

Thanks Michael,

yes, the data in the drive that I'm about to put in now is OK to be erased as it contains old server data.

[gregersenj]

Do not do this off-line.

Replace the disk on-line.
These boxes is engineered to Hot swap, hot add etc.
1. identify the failed disk.
2. removed the failed disk.
3. insert the spare disk.

If the spare disk is a used disk, with data on, the server still will rebuild the RAID on it.

Doing is off-line is a high risk, if you are not very sure, that you know what you're doing.
You need to be 100% sure, that the disk that you want to keep, got the latest timestamp, or that the metadata on the spare disk is cleared.

Keep it simple.

Br
/jag