Question Windows 10 with SDD as C, RAID 1 with a Smart Event starting it very slow

caliskier

Distinguished
Dec 29, 2012
93
1
18,545
Hi,

I have a windows 10 PC with a MSI motherboard that I set up two 2TB HDD in RAID 1 (Mirror) using the MSI onboard RAID controller, not a s Software RAID.

One of the disks says "Smart Event" in yellow on startup with the total raid "bootable" and in "rebuild" mode. I was able to get all my data backed up onto other drives.
Windows starts very very slowly even after Windows 10 refresh.

Question one, are the two issues related? Smart Event and slow windows 10 start?
Question two, its not very clear which disk is in "smart event" mode. How is that determined?

For Question 2, here is what I know:
Cant determine anything in the device manager, it just says you have a RAID 1 working fine.
In BIOS I cant tell anything.
On startup I see the raid volume is in rebuild status, and both disks and their model (models are identical) and serial numbers, and I can tell by serial number which has the smart event (maybe that serial number is printed on the drive and I can figure it out that way)
I also see an ID number which are 4 and 5, while drive numbers are 5 and 6.

Also as a 3rd piece of information. Before I wiped the machine it was freezing often, like every hour or if I tried to wake it up with the mouse it would wake up and I can see the screen, but no mouse movement, frozen, no control alt delete. When I went to pull the drives to protect them for the windows refresh I did not remember were the wires went (woops,) and plugged them back in wrong and crossed the raid drive with another 3TB drive I use for TV recording. I turned it on with them plugged in wrong and it said the status was degraded, I fixed this and now it says Smart Event on one of the drives as described above.
Note 1: All this happened BEFORE I wipped the C SSD for Windows 10 reinstall.
Note 2: Its possible and likely that I was having a "smart event" before I screwed up the wires on the HDDs. I saw yellow text when I would start the computer but did not pay attn. I would have expected it to be much more clear like "Hey dude! your RAID is Failing!!!" I know what to look for now.

Can anyone give me some guidance?
 
Last edited:
Follow-up, I already answered question 2, the S/N is on the drive itself. So now my question is, is the drive good or bad? as in did my miswiring cause the issue?
Secondly, is the fact that the RAID is attempting to rebuild and having a SMART event slowing windows startup down?

With the drives disconnected windows starts up quickly, but then of course is looking for those drives.
 
Yes, the RAID rebuilding itself will cause slowdown.
SMART Event refers to one or both drives dying.
One of them needs to be replaced.

As far as 4-5 and 5-6....some things start labeling at 0, some things the labels start at 1.

Thanks brother, so my mis-wiring is irrelevant to the problem?

I guess what I am asking is this: Only when I had them hooked up wrong did I pay attn to the message at start-up. It could have been there all along. When I turned it back on it said "degraded" under RAID status at startup so I knew I had them wired wrong, and I did with one crossed with the 3TB DVR drive. Now the wiring is fixed and it says "rebuild" Should it quickly go back to normal because the data should not have changed, or would the RAID controller get confused and think it needs to rebuild itself all over again?

I am asking as I know RAID 1s of this size take over 2 days to rebuild according to what I have researched.

Thanks,
 
If I may come at you again, i am not being clear. You are helping a bit of a novice, thanks for you patience.

When you said "And adding a whole different drive (the 3TB) WILL cause the whole array to rebuild. " that made me think that I did this and it can be fixed if I give it time...

I have Two 2 TB drives in RAID and 1 3TB drive not part of the RAID, and then the C drive SSD for the operating system
One of the 2TB drives gets crossed with the 3TB drive for one or two quick power on cycles, so the 3TB drive is on the RAID port with one of the 2TB drive, completely different data.
I switch the wiring back such that the 2TB drives are back on ports 5 and 6 (the RAID ports). (I am not sure if the wiring within the RAID itself is what it was (i.e. ports 5 and 6 could now crossed w/in the raid), but I hope that does not matter as I have no way of knowing)

So the ultimate question is, did those actions screw me up enough that it will take 2-3 days of the machine running for the raid to rebuild itself with my existing 2TB disks (I.E. there is no bad disk, just a screwed up raid that needs time to repair?

or...

Do I have a busted drive that is "trying" to rebuild itself and I will never get there and I just need to buy a drive now?

Thanks again for your help, its very appreciated.
 
Here we see one of the issues with a supposedly 'easy' RAID 1.
Unless you know why you need to, and know what you're doing with it...it quickly becomes more of a problem than what it was trying to prevent.

So.
2x2TB in RAID 1
There was a "SMART Event" ? That indicates a dying/dead drive.
Then, replace one of the 2TB with the 3TB.
This will cause the array to start a rebuild.
Then, take out the 3TB and replace it with the original 2TB.
This will again cause it to try to rebuild. The unremoved 2TB was being rebuilt with the 3TB.

Now...is one of the 2TB faulty? Unknown.
What specifically led to you seeing that "SMART Error"?


Lastly....(and I always ask this) Why the RAID 1?
What were you hoping it would do for you?
 
You are awesome, thx!

Here we see one of the issues with a supposedly 'easy' RAID 1.
Unless you know why you need to, and know what you're doing with it...it quickly becomes more of a problem than what it was trying to prevent.

Word! I screwed up when I did not look at the message and then crossed the wires

So.
2x2TB in RAID 1 -Yes
There was a "SMART Event" ? -Maybe, I did not pay attn to the message until I crossed the wires, but I do remmber seeing that it was yellow, just never read it. That indicates a dying/dead drive. - Got it
Then, replace one of the 2TB with the 3TB. - Yes, that happened
This will cause the array to start a rebuild. - understood
Then, take out the 3TB and replace it with the original 2TB. Yes, and maybe not wired exactly like it was within the raid before, 50/50 chance.
This will again cause it to try to rebuild. The unremoved 2TB was being rebuilt with the 3TB. - Ok, understood

Now...is one of the 2TB faulty? Unknown. - right! and, do I need to wait 2 to 3 days to find out? I can do that my stuff is backed up, just annoying.
What specifically led to you seeing that "SMART Error"? I saw it right after this step you detailed above "Then, take out the 3TB and replace it with the original 2TB."


Lastly....(and I always ask this) Why the RAID 1?
What were you hoping it would do for you?

All my data and wife's pictures are on the drive, if one disk fails I want to be able to swap out a new one and rebuild the raid. I also have a third drive I use to backup the raid separate but I am not as studious about backing it up.
 
I fear you'll have to let it finish, and then see if it reports one or the other as bad.
Then, replace THAT one and let it rebuild itself again.


And for the other question...
Given a good backup routine, a RAID 1 is not nearly the data protection thing you wish it to be.
It is good for drive redundancy, not data redundancy.
Corruption, ransomware, accidental deletion...RAID 1 does nothing for that.

Backups do not need to be a manual process. Many of the current tools have full and easy automation.
Set it and forget it (until you need it).
There is no need to be "studious".
I use Macrium Reflect.
Every system, every drive...backed up to another drive, every night.
In my case, a Qnap NAS box. Could just as easily be an external or other drive.
There is no reason to do that manually.


That whole procedure justified itself with a single drive fail.
One of my SSD's died, 605GB data on it. Motly photos.
Slot ina new drive, click click...all 605GB recovered exactly as it was at 4AM that morning, when the nightly Incremental was run on that drive.
No RAID 1 needed or wanted.
 
I fear you'll have to let it finish, and then see if it reports one or the other as bad.
Then, replace THAT one and let it rebuild itself again.

Ok, I will do that and give it to like Monday. If Monday night it still is messed up I will know.

And for the other question...
Given a good backup routine, a RAID 1 is not nearly the data protection thing you wish it to be.
It is good for drive redundancy, not data redundancy.
Corruption, ransomware, accidental deletion...RAID 1 does nothing for that.
Truth!

Backups do not need to be a manual process. Many of the current tools have full and easy automation.
Set it and forget it (until you need it).
There is no need to be "studious".
I use Macrium Reflect.
Every system, every drive...backed up to another drive, every night.
In my case, a Qnap NAS box. Could just as easily be an external or other drive.
There is no reason to do that manually.
Is your NAS box in a RAID?


That whole procedure justified itself with a single drive fail.
One of my SSD's died, 605GB data on it. Motly photos.
Slot ina new drive, click click...all 605GB recovered exactly as it was at 4AM that morning, when the nightly Incremental was run on that drive.
No RAID 1 needed or wanted.

Picking up what you are throwing down. So, just keep the good drive, remove the raid and instead of buying a new drive, hook up a backup drive of some kind and have it refresh sometime each day? Could that be an internal drive? The replacement for this is $40. Maybe I put the same drive back in there (working one), UNRAIDed, and then can windows or something else do a procedure to replace updated files and folders to a second HDD daily?
 
Currently, the QNAP Nas is NOT a RAID.
Initially, I had it as a RAID 5, just for experimentation.
That has been undone.

For your backups...
Doing it to another internal is OK, but it ALSO needs to go to an external something.
An internal drive is subject to the same data faults as your main drive is.
Get some nasty virus or ransomware....all is toast.
Some external thing that is physically OFF will not be affected.

As above, Macrium Reflect will do that backup thing on whatever schedule you set up.
 
OP, go and download Hard disk Sentinel https://www.hdsentinel.com
Its the De-Facto best software to analyze and fix HDD issues, it checks health, smart, errors, temps and more durng day to day or as run tests.
It has 1 month free, I dont have RAID but I always have bunch of large HDD's, so eventually i paid for a full version, but for you, even 1 month is enough to see whats going on.
Install it, and load it up, if your hard drives have mechanical failure or bad blocks, something discoverable by the HDD firmware it will say so in the program.
On top of program tests, it allows to run in-build Self Tests that every HDD has build-in on its firmware, it has 2 tests, short 2 minute test that tests all the essentials and long full disk tests that tests for everytng, its hardware based tests run onby the HDD itself, and since your HDD's are so small, it wont take long, maybe 3-5 hours for complete test per HDD



P.S. if you backed the RAID, my recommendation will be to disable it, check both HDD's, run full tests on each, see if it discovered something.
Then you should run a complete Write tests on each HDD, it writes random numbers from start to finish and then checks if they can be read from the HDDs, if it has bad block it marks them, and replaces the bad blocks with clean blocks from the pool [each HDD had a number of blocks in the pool to replace the bad ones].

BUT if it indeed finds bad blocks and fixes them, DO NOT use such HDD for RAID, its ok to use for storage if you keep an eye and there are no more new bad blocks after the fix, but its NOT good for RAID 100000%
 
Thanks both, stopping for some dinner and some starwars for my 6yo. A new hope, has not seen it yet, will get on this. really appreciate the love today fellas. Will get back to you on what I do and if I have addnl qs.
 
So in between dinner and Starwars I looked up Macrium Reflect. WOW, looks like the free version has what I need. Does differential backup on daily refresh which would just replace what changed. Does it just look for a date stamp to see what changed?

This is really cool! So what prevents a virus or malware from getting copies over to it? how does it lock it self out and off?

Can any drive work, say this one? WD 2TB Elements Portable External Hard Drive
 
Currently, the QNAP Nas is NOT a RAID.
Initially, I had it as a RAID 5, just for experimentation.
That has been undone.

For your backups...
Doing it to another internal is OK, but it ALSO needs to go to an external something.
An internal drive is subject to the same data faults as your main drive is.
Get some nasty virus or ransomware....all is toast.
Some external thing that is physically OFF will not be affected.

As above, Macrium Reflect will do that backup thing on whatever schedule you set up.

I am ready to do the Marcium Reflect, my system is running a lot smoother at least without me restarting and rechecking, still going to wait until Monday to restart it and see if the RAID finished.

What HDD should I use for Marcium Reflect that is physically "OFF" like you stated?
 
Differential is ALL of the changes since the last Full backup. Over time, the size of the next Differential is larger.

Since you'll be doing a series of these, you can go back in time to any previous backup before the BadThing happened.
If a drive is OFF or disconnected, any malware can't access it.

For a drive that you want OFF, an external would be indicated. Power off or disconnect, until the next Full or Differential.
 
Can the on/off be automated or is that a manual thing? I was excited by your comment that its all automated at 4:00am.
The backup pushed from the PC at 4 AM is automated.
That happens nightly, for all drives and all systems. Each drive gets its own time slot. 12:00, 12:30, 1:00, etc.

The drive and location that is mostly OFF...that happens weekly. That is the next level backup.
I just need to turn that on at the appointed time on the appointed day.
 
So, in the highly unlikely event that my PC, and the nightly backup, are both totally corrupted, ransomed, whatever...I can power up the offline drive, and go back to whenever that was created.
Unlikely that I would fail to notice a major data fail and have that leak into the offline backup.

Of course, for truly life changing things, those are stored in a drive that is actually offsite. Refreshed whenever needed.
COVID19 is making that procedure a bit weird.
 
For instance, this is the folder tree on my storage device, with the folders for my main system.
Each drive gets its own folder. In that folder are the Full and Differential/Incremental images.
7775VWa.png


Notice the Date modified time stamps.
All this AM, between midnight and 3 AM.
All automated.
 
You are awesome, I think I am going to switch to this.
1 HDD for the Safe deposit box, annual backup.
1 nightly HDD backup, that will be internal as it will be the other RAID disk if I can show it good,
And finally one, weekly off/on drive, I can use my phone as a reminder or just do it when we take the trash cans to the street and pick them up.

What would you recommend as that 3rd drive, the weekly on/off drive? Actually thats pretty simple right, that could just be an ordinary drive like this WD 2TB Elements Portable External Hard Drive, right? I would rather just plug it in and press a button and it does the differential thing for the weekly on/off. Is that simple to set up?