Save to swap disk on a DELL Priolant DL380 Gen9?

[1337guru]

Hi all,

Hopefully someone can give me answer I have a disk error in the SMART Array Controller.

Status: Error
Capacity: 600 GB
Model: HP EG000600JWEBH
Serial Number: Confidential info erased. 
SAS HotPlug Supported: Yes 
Speed: 10500 RPM Drive Configuration:
Data Sanitize Erase Support: Not Available

The disk have a amber solid light, i found the meaning of this color:

Amber = Drive failure, link failure, ormismatched configuration.

So far so good, the other disks ( raid 5 ) have a red light button with the "do not remove icon" i did some research and that means that you cannot remove one of these disk or all your data is lost right?

I orderd the exact same disk for replacement.

My question is: Can i safely remove the failed disk en swap it with the a new disk ( hot swap ) ? the raid session will be restored automatically?

It is a bit uregent because a important VM is running on it and if one more disk failed, i losing all of my data.

Any help would be gratefull!

[BPSingh · Accepted solution]

Greetings!

The other disks ( raid 5 ) have a red light button with the "do not remove icon" i did some research and that means that you cannot remove one of these disk or all your data is lost right?

-> Yes, RAID 5 has fault tolerance of one drive. In "this" case since one drive has already failed, it's by design to see "do not remove LED" on the other drives because pulling any of these 'healthy' drives would result in data loss. 

I assume that the drives that you have are hot plug drives. If they are, then you can hot swap (replace) the failed drive and the RAID rebuild should start automatically after the replacement. 

[1337guru]

Thank you for your reply! I will replace the disk tommorow.

[1337guru]

@BPSingh 

Hi sire,

Yesterday i swapped the disk with a new one and the rebuilding of the RAID 5 has been succesfull. But this morning i checked it again en the disk is already in "Predictive Failure" state. How is that possible after 24 hours? Is this just bad luck with the new disk or can it be something else? Other question, do i need to wait when it goes to "failed" status or is it smart to replace this disk already with a new one? 

Any help would be great.

Thank you

[BPSingh]

Greetings!

It's unlikely that replacement drive will fail again within 24 hours. However, you may try to replace it again and check. If the issue persists, then there would be some underlyng issue with the bay which needs further investigation. Please log a support case if the issue persists.

[1337guru]

Thank you for your reply.

I will swap the predictive failure disk next week. 

See what happends then, i let you know the result

[1337guru]

@BPSingh 

Hi Sir,

For now i leave it in the predictive failed mode, because everything is working fine and de RAID session is still good. If de disk turns into a failed state, than i will swap it Can take days, hours, or months right?

One more question: I see this errors on the predictive failed disk:

Hard Read Erros: 298

Recovery Read Errors: 666

Is that a lot? Is there a default for what is good / bad? The errors are not counting though, so i seems to be stable.

thank you

[BPSingh]

Greetings!

If the Hard read/write errors exceed the threshold then the drive will fail. Having said that, yes you can continue to use the server while the drive reports predictive failure.

Also, ensure the drive, controller firmware and driver are up to date. Please take a data backup before upgrading the firmware.