My understanding of RAID1 is that you have a minimum of 2 disks of same size. If one of the 2 fails, you remove it, replace it and the system creates a new mirror of the remaining disk on the new drive.
The failing or faulty drive is no longer in the system in order for data to get copied over to the new drive.

 

My understanding of RAID1 is that you have a minimum of 2 disks of same size. If one of the 2 fails, you remove it, replace it and the system creates a new mirror of the remaining disk on the new drive.
The failing or faulty drive is no longer in the system in order for data to get copied over to the new drive.

Thanks, but this doesn't appear to me to answer the 2 questions I have or provide details of the procedure. The procedure is key.

 

I don't know about your computer, but my Synology box has gotten several new drives over the years. When one fails, I just pop another one in, the drive gets formatted, rebuilt, and the RAID array is back to 100%. That's the way it "should" work on most RAID1 mirror arrays.

 

To Add to what is being said, You may need to boot into the RAID controllers Menu at bootup before Windows loads, by pressing the key combo shown on the screen to view the RAID configuration. If needed, you can remove the failed drive in the RAID configuration here, install new drive, back in the menu press Rebuild Array.
As stated,. with RAID 1 it should rebuild it automatically.

 

To Answer your questions in OP Post
1. Only you will know if this is a Hardware RAID controller or if it is a Software RAID, you are the one that set it up?! If unsure, this link gives you a command to type in to the Command Prompt run as Admin, to find out Technical Tip: How to identify hardware or software RAID in FortiAnalyzer or FortiManager

If you have a Hardware RAID, You will get a message after POST when the computer is booting for a split second to Enter the RAID controller Menu with a Key Combo, if you ignore it, it will boot directly into Windows. As stated in post #4. You should Boot into your RAID controller View the RAID Array. It should tell you one of the drives is bad. You would want to choose to Remove this drive from the Array. Shut down the computer, and replace the bad drive with the good one If it is the same capacity size.
On Bootup, the RAID array should rebuild. If not, boot into the RAID Menu, View the Array and it should recognize the new drive. Press Rebuild Array.

2. RAID 1 is a Mirror of the original drive. So, if you have 2 of the exact same size drive, Drive A and Drive B. If Drive B fails, when you replace it. The Array rebuilds using Drive A's data to make a Mirror copy again.
RAID 1 arrays only show as one drive in Windows Explorer. To see if the array is working, you have to boot into the RAID Menu. Or in the Bios.

 

As stated,. with RAID 1 it should rebuild it automatically.

Aye, there's the rub: as stated by whom? An ASUS technical support person? Somebody with reproducible experience of RAID 1 on this specific mobo?

Although there is no information on the ASUS web-site that I can find about disk replacement, there are some very abbreviated notes in the mobo handbook which are greatly concerning me: firstly to build a RAID (and presumably rebuild it) I am required to set the connection mode on those drives from AHCI to RAID. The BIOS screen warns me that all data on these drives will be lost if I do this; if I don't do it then I cannot rebuild the RAID. I find this puzzling, as I have been running this RAID on this specific mobo for some years, apparently in AHCI mode.

Secondly the mobo manual and the BIOS both tell me that if I delete a disk in a RAID 1 then all data is lost - on both drives, by implication - and further that the rebuild of the array requires that the 'disk' be formatted - again with loss of all data.

You can see that I really need access to an authoritative procedure/discussion of drive replacement in a RAID, without loss of data, on this mobo. Where do I find that? Google doesn't find it.

 

Please attach a screenshot of Disk Management in your next post to see your RAID setup now.

 

As a technician, in your position, I would test, outside of windows, both drives individually to determine which drive is faulty and why.

I would then use clone software to create a copy of the good disk to a new drive.

FIRST IT WOULD BE A GOOD IDEA TO CREATE A BACKUP COPY OF THE GOOD DRIVE.
Consider also replacing both drives.
This means you have the opportunity to recover your system in the event that anything goes wrong, which includes the possibilty of power failure or the original drive going faulty too, since both drives are equally old and worked just as much.

 

1. Only you will know if this is a Hardware RAID controller or if it is a Software RAID, you are the one that set it up?! I

Sadly this is not true: I know as much now as when I first built this PC. It did not tell me then whether it was hardware or software raid so I had no way of knowing. I merely followed the prompts in the BIOS.

The link you directed me to suggests that I run a command (#diagnose) which is unknown to the CLI in my version of Win 10 - as is the command 'diagnose' or even '# diagnose'.

There is no option during POST to enter any sort of RAID administration - this is all done from within the normal BIOS interface, which is common to all models of ASUS mobo of this generation.

I am sure that were I to enter this part of the BIOS - which, as I have said requires me to switch my drives from AHCI to RAID, with the loss of all data, that the BIOS RAID services will NOT tell me which disk is #bad' because at this point it is not being detected as having failed. CrystalDiskInfo reports that caution is required in continued use of this disk because the count of alternate sectors which have been assigned is above a threshold. It is currently at about 384. I don;t know is this is an indication of imminent failure or not, since SEAGATE do not reveal what is the limit to the reallocated sector count for this particular model. What I do know is that it has not changed in the last few months, which is why I am in the planning stage for disk replacement (I have a suitable disk available) rather thanseeking to change it right this minute. The risk is that the BIOS RAID function will not give me any warning of imminent failure as CrystalDiskInfo is doing; it will tell me only after the fact, when the drive becomes unavailable.

 

As a technician, in your position, I would test, outside of windows, both drives individually to determine which drive is faulty and why.

I would then use clone software to create a copy of the good disk to a new drive.

FIRST IT WOULD BE A GOOD IDEA TO CREATE A BACKUP COPY OF THE GOOD DRIVE.
Consider also replacing both drives.
This means you have the opportunity to recover your system in the event that anything goes wrong, which includes the possibilty of power failure or the original drive going faulty too, since both drives are equally old and worked just as much.

CrystalDiskInfo has already told me which disk (by serial number) is potentially faulty - and why. This is not a disk failure condition yet, but will become so.

A clone copy of the good disk is of limited value: the disk contents are continually changing, but the 'single' RAID disk is backed up, twice, every hour or so to external larger USB attached drives. This is all good advice which I have practiced since my earliest experience of disaster recovery in the IT industry back in 1962. However none of this addresses my need and the reason for my original post: where do I find a definitive, proven procedure for replacing one the drives in my RAID configuration on my mobo without loss of data?

My apologies if I haven't made that need clear.

 

Please attach a screenshot of Disk Management in your next post to see your RAID setup now.

Screen shot attached (indirectly; hopefully). Drive E is the RAID drive

 

Maybe the OP can clear this up.
It looks like Disk 0 Nova D: and Disk 2 Nova E: are what you are trying to make a RAID array?
right now, these drives are separate disks and they are not in a RAID configuration.
To Make a RAID array, you have to install the Intel Management Engine Interface driver from ASUS download site for your model. PRIME H370-A - Support
Both Disks must be the same size capacity. Disk 0 Nova D: is 500 GB and Disk 2 Nova E: is 1TB.

Maybe this will help you? https://www.windowscentral.com/how-set-mirrored-volume-file-redundancy-windows-10

 

Maybe the OP can clear this up.

Yes, I can clear this up.

For some reason there is a discussion here about how to build a RAID. This is irrelevant: the RAID is already built and has been operating for some years. The screen shot of disk management, along with the comment in the post that contained that screen shot, explicitly states that the RAID is reported as Disk E. Disks C and D have no part in the RAID configuration. I do not understand how it could be assumed that I was trying to create a RAID with different sizes and different technology of drives.

The link to the www.windowscentral.com describes how to do disk mirroring in Windows - a feature which I decided years ago never to use. The article makes no mention of the process for replacing a disk in a RAID managed by BIOS and intel drivers on my specific mobo.

I am unable to understand why there is this level of confusion. I tried really hard to make my original post clear as to the reason for my posting to this forum. But my post appears to have created off-topic discussions. What should I have done differently?

 

There is no confusion on my part ... I was trying to give you a workaround, which you have already done in part by using a , new to me, disk diagnosis application that has warned you of smart problems and identified the possible failing disk.
I trust that your research is as you say, not giving the information that you seek. My only experience to date with RAID was with a striped version, ie RAID0 in a 2TB (2x 1TB DRIVES) which resulted in a lot of searching and understanding.
In servers, when 1 drive out of several goes down, one just (hot) replaces the drive and wait for the rebuild to finish, maybe after a day or two, depending upon the size.

Everybody here responding to your question is trying to push into a direction that will provide you with an answer or even a workaround since it is highly unlikely that anyone else will have exactly the same hardware and problem as you do. We search the internet for answers, since, as you must know with your years of experience, depending upon your search wording, different results will be shown. A lot of us have years of hands on experience in IT and services. When there is no direct answer we try to use that as a base for what "COULD" be done.
Advice is freely given and you are free to act on it or disregard it. With your experience and research you have no answers. You are asking for some ideas about how something could be done, hoping that someone here will have that one answer that will solve your problem that not even the hardware support people of your devices have been able to solve.

Currently, BOTH DRIVES ARE FUNCTIONAL in spite of an increasing number of SMART interactions that tell you that one of the drives is failing.

The data is intact and to all intents and purposes, trustworthy.
Cloning each drive to a new drive should allow that new drive to replace a failing drive.
You lose nothing by trying this except time, you have a backup already. I do not know if there are differences somewhere in the MBR or FAT that distinguish the drives between Master and Copy, which is why I say to CLONE not COPY each drive to a new drive. This should maintain your data integrity.
Alternatively since you have a back up, replace the faulty drive and wait for the hatdware / software to kick in and advise you about rebuilding.

 

Further to above, another workaround would be to replace both drives. Create a new (bootable?) RAID1 system, then copy your back up or good drive data to the new clone.
I am not exactly sure whether this might work since if the RAID is bootable and into Windows, windows will want directions to find the bootable partition.
This is outside my realm of expertise.
Theoretically, creating a RAID1 with 2 new disks, then reinstalling windows and copying your data to the drive should work.
Only trying it will show.

 

Sorry, please read my original post. I don't need a work around; I don't need to replace both drives; I don't need a bootable RAID. I don't need off-topic opinions.

What I do need as a reference to a document which gives adequate, authoritative, technically valid information on the recommended procedure to replace one of the 2 drives in my RAID, on my specific motherboard, without loss of data.

 

Open an elevated powershell prompt and copy/paste this Get-VirtualDisk hit enter.

What is the output?

 

Simply: nothing. Here's the snip:

 

Very simply, Hardware RAID uses a separate controller card that's installed in a computer which manages the array and communication with the drives. What is considered "software RAID" is software like Intel's that's built into the motherboard, and it works very well. It's called "software" RAID because it does not require any additional hardware.

When you have a failed drive in a RAID1 array, you'll replace the drive, boot up into the RAID management software (usually Ctrl+I), add the new drive to the array then select the option to rebuild. That's it. There's far too much other bad info in this thread.

 

@Fred Garvin
Where do we see the drives are in a RAID array?

 

Thats because you have no drives in RAID. If you did, there would be output.

 

That command has nothing to do with detecting an Intel RAID array.

 

the Windows Central Link in Post #13 near the bottom shows what 2 disks in a RAID configuration would look like. The Disks should be Dynamic, and show an umber colored toolbar, with Layout being Mirror.
Your drives do not show as being in a RAID, and the PowerShell results show there is no RAID configuration on your computer.

For more proof
You can also open a Command Prompt as Admin
and type Diskpart and press Enter. then type
List Disk and press enter. This will list the disks by their number. then type
Select Disk # (where # is the number of the disk you want to select).
This will tell you the type of disk it is and if it is part of a RAID or not

How do I check Windows RAID status from the command line? - Microsoft Q&A

Hi All, Windows 11 - 22H2 How do I check Windows RAID status from the command line? This is what I am thinking of doing: @Echo off ( echo List volume echo exit ) | C:\Windows\System32\diskpart.exe And then checking to see if Status=Healthy for…

learn.microsoft.com

 

I am not offering workarounds or asking your to replace drives.
Based on the information you have provided, you do not currently have a RAID array with your drives on this computer. You can confirm this by opening a Command Prompt as Admin and typing Diskpart.

 

Diskpart differs with your assertion:

 

TL;DR
You can view the drives, rebuild an array and see the status of an Intel RAID array within Windows if you install the Intel RST management software. I believe Intel's naming of the software has changed and may vary for newer or older controllers. Access to an Intel RAID array outside of Windows varies with a computer's BIOS, check the Mobo manual. You may be able to do it from within BIOS or watch the POST screen for a command to enter, like CTRL+I which will take you to the setup software that shows everything.

No, you don't have to delete a RAID1 array in order to replace a drive in it. That defeats the whole purpose. You need to check your motherboard's manual for the steps or use the Intel RST software. Intel has several different RAID products and so the steps vary.

 

I am bowing out of this discussion.

It would appear that, in spite of thoroughly researching the problem yourself and finding no approprate answer, you insist that there has to be an answer but only in the format that you want.

My late mother used to say "There are a thousand ways to skin a cat!" and I have found that to be very true during my career in electronics and PC repairs, also life!

Good Luck trying to find your needle in the haystack.
These days when you lose a needle, it's easier to get a new one!

 

use the Intel RST software

Yeah, that worked - thanks. No loss of data, just loss of time: rebuild took a few hours. Problem closed .... almost:

Turns out the the drive reported by CrystalDiskInfo as 'good' was the one causing all my problems: it has a defective SATA connector causing the Intel RST firmware to intermittently not be able to 'see' the drive - leading to repeated attempts to rebuild the array. Bottom line: working raid, still containing a drive which has a (stable) sector reallocation issue.

 

You need one good drive to rebuild the array. If you think there's an issue with both, run a backup on the pair or just the solo, better drive then replace them. Storage is cheap.

 

I would recommend that you get an external drive, larger than your RAID drive, and make a disk image of your current drive. Macrium Reflect is a good program and not terribly expensive. That has nothing to do with rebuilding a RAID array, but it would remove the worry of data loss.

 

Would you have made this recommendation if you had fully understood my original post?

 

To safely replace a failing HDD in a RAID 1 setup on your ASUS Prime H370-A motherboard, follow these steps: First, verify that your RAID 1 is managed through the Intel Rapid Storage Technology (RST) in the BIOS, which means it's a combination of BIOS and software management rather than a pure hardware RAID.

To replace the failing drive, power down your system, remove the faulty HDD, and install the new drive. After rebooting, enter the BIOS, go to the Intel RST section, and confirm that the RAID 1 array recognizes the new drive. The system should automatically rebuild the array using data from the healthy drive. Ensure the replacement drive is blank or initialized with no data; the rebuild will always copy data from the remaining good drive to the new one, not vice versa. Monitor the rebuild process through Intel RST or Windows Disk Management to ensure data integrity.