[Resolved] Faulty RAID1 disk now shows as foreign

Hello Dylan,

Many thanks for your reply, much appreciated.

It's a PowerEdge R320 with a PERC H310 Mini (Embedded). Properties listed below, it would be great if you could confirm 100% that this is a hot-swappable model, from what you've mentioned it sounds like it.

To confirm, I just need to swap the faulty 'foreign' HDD and insert the new one (which is identical model), I can do this while the server is online and it should start rebuilding?

I currently see in the option you mentioned (Storage -> Virtual Disks) on the RAID that is currently faulty it is labelled 'Degraded' and in the option menu there is a 'Replace Member Disk'. I assuem I'd need to action that if it does not start rebuilding automatically?

Where would I see a 'rebuilding' message? In the 'Status' field under the PERC H310 Mini (Embedded) page?

Thanks again for your help.

Controller Information PERC H310 Mini

ID 0
Status
Name PERC H310 Mini
Slot ID Embedded
State Degraded
Firmware Version 20.12.1-0002
Minimum Required Firmware Version 20.13.1-0002
Driver Version 6.600.21.08
Minimum Required Driver Version 6.803.21.00
Number of Connectors 2
Rebuild Rate 30%
BGI Rate 30%
Check Consistency Rate 30%
Reconstruct Rate 30%
Abort Check Consistency on Error Disabled
Allow Revertible Hot Spare and Replace Member Enabled
Auto Replace Member on Predictive Failure Disabled
Persistent Hot Spare Disabled
Cache Memory Size 0 MB
Spin Down Unconfigured Drives Disabled
Spin Down Hot Spares Disabled
T10 Protection Information Capable No

Hello,

Replacing the drive would likely be for the best. As for your question about it being foreign, yes, you can swap it live, provided it is running through a backplane. However, I'm fairly certain the R320 with a PERC included a backplane by default. You can swap it live because a drive in foreign status is not an active array member. Generally speaking, all you need to do is swap the bad drive with a good one. This should automatically kick off a rebuild. However, I would still double check it in OpenManage. You can check on the rebuild under Storage > Virtual Disks. If the VD isn't rebuilding, there is a drop down menu where you can choose to either "Replace Member Disk" or assign a hot spare. When I do this, I normally assign the replacement as a hot spare and let the controller handle everything.

If the hard drives are accessible from the front and without removing the side panel, you'll have a backplane. This output doesn't tell us, unfortunately. Your steps are also correct. My expectation is that you'll see the PERC handle everything, but you can verify by watching where it shows your array as "degraded." It'll change to "rebuilding" once it starts incorporating the new drive.

Hello @Dell-DylanJ 

Thank you for your reply.

I have done this (removed drive hot swap) and inserted a new replacement (identical make/model) and all went well, but it is now stuck at 0% rebuilding and isn't moving.

What would be the correct action to take now to reattach it to the RAID1 alongside the other disk.

Maybe I am being impatient, but it hasn't moved from 0% in over 30 minutes so it looks like it isn't doing anything (its a 4TB array).

What would be the correct steps to force it to start and can this be done via OMSA?

Thank you!

Looks like I was being impatient, it is now rebuilding, currently at 5%.

Will it attach itself as it was before or will I have to take any action when it reaches 100%?

Thanks

Once 100% is reached, it should join the array on its own.

that HDD will fail , S.M.A.R.T test on my LAB PC,  with HPA raid card and running HDsentinal pro. w10 box.

If yes, it probably has UNRECOVERABLE errors and if yes, and the meta data Raid is scrambled, then sure

that is now a foreign drive, for ever.  The weak drive is one that has run  out of spare sectors(seagate magic)

in hot swap bay , pull it, live,  if the drive is dead cold it is dead. if nice and warm well the motor works, grin.

plug in a new dell drive  same sized. (closer to same is always better) smaller never.

it will stay online (production) and runs slower, until it rebuilds.  the OSMA will say on the 1st screen

something like that and  xx % , hit Refresh from time to time if want most accurate, % , impatient like me.

this is the beauty  of real raid,  also tag the Sever with (what raid chip is inside)for next time.

what  PERC  is there. Knowing what hardware is in the box is allows SUPER GREAT to know.

To post the end result here, the steps I ultimately took were:

• Remove faulty drive while server powered
• Repalce with identical drive
• Took over an our to move from 0% rebuilding
• Took 4 days to finish rebuilding (4TB)
• Replacement drive automatically rejoined the RAID1 array

All good! Thanks to all for your help!