PERC S100 - RAID 1 Virtual Array delete or not

Hi, rebuilding occurs after a disk replacement or a disk fail. Which one was it, could you tell us? Also could you give us the part number for the disk (two if replacement)?

Hi

thanks for the reply

The disk failed and I then managed to get hold of a new disk with same Dell part number as the 0:0 disk ( they all match)

Will get part number when can access server later today.

Best

Marc

Above gives you the parts number

Above shows the disk stalled at 52% (and the other Array on this server)

Hello @Exminiman,

Thank you! I can see the part number of the disk. It is 1KWKJ and I can confirm the Disk is compatible with the server PowerEdge T310. Are you sure you replaced the failed disks and both disks are good now? If the server was on warranty, I would be sure to check the hardware logs.

How did you handle the process of disk replacement? Did you assigned the new driver as hot spare? I suggest you to check this post: https://dell.to/3o5OGoV

Regards.

Hi 

thanks for the link etc, I had followed the procedure listed below. 

In an effort to move things along I have run the process above again. 

Powered off the server ( no option to hot swap on PERC S100)

Removed disk 0:1 cleaned it with Diskpart.

Powered on server

 Checked BIOS RAID ( ctrl R), could do nothing as When you select Phsical Disk 0:1 it says still part of Virtual Disk

Let boot to OS

followed procedure in Open Manage

Disk rebuilt to 52% again, and stopped!

One thing I did notice was when a selected the HOT Spare to rebuild there was a warning about the disk not being large enough, or something like that ( but it is) This I took to be a message associated with the other Virtual  disk which is larger.

However, when I took over this Server a previous attempt had been made to rebuild the array using a larger disk, which had also stalled, at 54% I think from memory.

Is it possible that the Virtual Disk has  a rogue tag left over from this larger disk ?

Re an earlier question; is it actually possible to delete the Virtual Disk, preserve the data and rebuild with PERC S100?   The links in my first question to Dell instructions seem to say that you have to delete the Virtual Disk to rebuild it 

thanks for you help 

regards

Marc

Another thought

Does the RAID controller need to think that this disk has been degraded, rather than just stalling at 52%.

if so could it work to ;

power down server 

remove disk 0:1

reboot server to OS, without disk

power down server

replace disk

reboot server

And then go through rebuilding procedure….

I am not near the server so I am trying to formulate a plan for when i visit site

Hello again!

Honestly, I am starting to think there is a problem with the replacement disk or with the disk that stayed on the RAID. It is not normal that the rebuild stops at 52%. A PERC log analysis would be required. There are probably errors recorded on the PERC, either during the rebuild or in the historic log.

Do you have the possibility to make a whole backup, delete the VD, initialize to erase all data and the recreate the RAID and restore the backup?

Regards.

Hello Marc,

The S100 does not keep a controller log. You can find storage evens in the Windows System Event Log and also look in the LifeCycle Controller log. ( I think that is available in OpenManage. If not you can reboot and do F10 for the LifeCycle controller to view the LCC log.)

Hi 
how do I get a log analysis for a PERC S100?
best

Marc

Hi 

Could some one bottom this Dell "Help Article" out for me?

I keep going back to it, because it seems to say that you have to delete the original virtual signature before adding a new disk - am i misunderstanding ?

Please read the steps on the page/link and comment - even if its to say I am being stupid 

link https://www.dell.com/support/kbdoc/en-uk/000134482/perc-s100-s110-s300-how-to-replace-a-failed-hard-drive-on-a-software-raid

Heya Exminiman,

Apologies if this seems a little silly, but I'm not sure what you mean when you ask to bottom the article out. If you could clarify, I may be able to better answer this specific part.

Regarding the S100 controller, the controller is unfortunately not as robust as the hardware offerings. The S100 does not maintain a log, so troubleshooting issues with it can be troublesome, at times. Newer versions of OpenManage can sometimes capture helpful data in the Alert Log. I forget the exact text of the error, but the Alert Log was the only place to locate the message, so I would encourage you to review that log, if you haven't already. When there are errors with the RAID metadata, the array is not capable of rebuilding because the RAID algorithm would be missing the necessary data to complete the rebuild. This ends up necessitating redeployment.

From the thread, it sounds like your array may well be compromised. If you'd like to send a copy of the Alert log to me, I'd be happy to look it over for you.

HuI Dylan

Thanks for the reply, by "Bottom Out" I just mean, could some one confirm whether I am understanding the Dell Article is correctly and it says you have to delete the Array (or signature) if you need to add a new disk. ?

I should be able to get copy of Alert log across in next hour....

Thanks for your support

Best

Marc