• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Possible RAID failure- am I ****ed?

P

peppergomez

2[H]4U
2FA
Joined
Sep 15, 2011
Messages
3,423
I might be encountering a problematic ProRAID (http://www.mediasonic.ca/product.php?id=1358153144) disk failure issue.

A friend set it up for me. I think it might be in RAID 0, as the drives are seen by Windows 10 as a single drive.

One of the discs (not sure which) makes a clicking sound upon booting up/turning on/spinning up. This sound stops after maybe ten seconds. The drive isn't recognized by Windows and there is an error light on the front of the RAID case.

I haven't been running it since this happened.

There wasn't any warning or indicator of odd behavior. It just dropped out in the middle of a session.

Any suggestions?

Thanks.
 
How many drives do you have in the enclosure? How large are each of the drives, and how large was the single drive as it was presented to windows? If it were the total size of all the drives, it was R0. How important was the data on the array and do you have a current backup?
 
4 drives. 2TB each I believe, and 8TB. So yes, R0 I think.

Pretty important. I back up every few months but there is some data that I haven't backed up.
 
Well, professional recovery (drivesavers, etc) would be your next step, but for RAID is most likely a multi-thousand dollar proposition. If that is not an option, the next step would be to try and make a clone of the failed drive. Maybe it is a failed slot in the enclosure and not a failed drive (possible but NOT probable). Maybe you can get it to stay up long enough to make a clone and then do a RAID recovery. Keep in mind any operation you perform on the drive could possibly damage it further. In the end, regardless of the outcome RAID is NOT a backup, and you should get yourself an 8TB drive (or two) and keep staggered, historical backups of your important data.
 
Thanks. Any particular backup approaches that you recommend, or would manual approaches be sufficient.

Also, do folks have any recommendations for reasonably priced online backup options? I am a photographer and have 10s of thousands of large RAW files (50MB each). So we're talking 4 or 5 terabytes probably.
 
There are a number of free and paid backup apps, both open and closed source. For free AOMEI, Veeam, Macrium etc. For online, backblaze is one of the better options for the money.
 
PS There are only a few folders of data on the failed RAID that I haven't backed up. I don;t suppose there would be any ways of finding this data without having to run the drive? Nah, I guess not...
 
Not in RAID 0 you can't. RAID 0 data is not recoverable without all drives working due to the way files are split across multiple drives. It is also for this reason that running RAID 0 for important data is not recommended unless it is backed up at least weekly, and highly recommended to do daily backups for RAID 0.

However, there is a chance that it is set up as JBOD, and in the case of JBOD, the data on each disk is recoverable without the other drives. The data on the failed disk wouldn't be recoverable if the failed disk can't be read though in JBOD.
 
Your friend doesn't know what he's doing if he knowingly let you put important data on there.

That enclosure supports RAID5 and 10, either of which would have prevented data loss in this situation.
 
yeah +1 on the friend being either stupid or mean to put you in this very easily preventable situation.

The clicking noise is mechanical failure, and outside of paying a LOT of money to clean room transfer the platter, you are 99% hosed.

One last ditch effort you can try that some people have had success with - google "freezer trick hard drive" - some people swear by it, others call BS. When you've reached that point, what do you have to lose really?
 
My bad folks.,..it's RAID 5. I'm a noob and didn't know at first.

The model is Media Sonic ProRAID HFR2-SU3S2. It's a 4-bay HDD.

The RAID isn't recognized by Windows in Device Manager (either via USB or via SATA).

The xH error light is on, but the Rebuild light is not.

Is it possible to determine which of the four drives is faulty? And would it be possible to pull this drive out and put it back in, or put a new blank drive of the same model, so that the RAID rebuilds itself?

Any other assistance or ideas would be appreciated.

Thanks.
 
Well, if it was 8TB advertised as you said above, and you only have 2TB drives then it is NOT R5. If it were R5, and one drive failed the array would be in degraded mode but still accessible. I just downloaded the manual, on the front of the panel near the top, there are LED's for 0, 0, 1, 3, 5 & 10. Which is illuminated? Next, just below that are the fan lights, and then below that are the disk lights. Which are illuminated, and for each are they blue, purple or red? Next, there is an LED to the left of each drive, that is the error light.. Which one is illuminated?
 
mwroobel said:
Well, professional recovery (drivesavers, etc) would be your next step, but for RAID is most likely a multi-thousand dollar proposition. If that is not an option, the next step would be to try and make a clone of the failed drive. Maybe it is a failed slot in the enclosure and not a failed drive (possible but NOT probable). Maybe you can get it to stay up long enough to make a clone and then do a RAID recovery. Keep in mind any operation you perform on the drive could possibly damage it further. In the end, regardless of the outcome RAID is NOT a backup, and you should get yourself an 8TB drive (or two) and keep staggered, historical backups of your important data.
Click to expand...

One thing the OP can try is attempt to purchase an identical model & revision drive off of ebay and replace the controller board on the drive. A lot of professional recovery places do exactly that before going onto more drastic steps.

Anyway, I had a clicking drive in a raid, and had some spare drives. Turns out that replacing the controller board revived the old drive, while the bad controller board exhibited the same clicking symptom on the non-symptomatic drive.

I believe the clicking is a safety mechanism in action for parking the head as fast as possible before the head would touch the platters and cause damage. Usually if some capacitor or other electrical component on the controller board has failed and can't get the platters spinning fast enough, it doesn't levitate the head enough (only a couple of microns) and the head is put back into the parking spot.
 
Here is the device that I have:

http://www.mediasonic.ca/files/1383763501_0.pdf

In case these details are useful:

I am using Seagate Barracuda 3TB drives, model ST3000DM001

12TB total, but I think 9TB was accessible for storage.

I re-seated the drives, and booted the RAID up. Now there are two red Error LED lights illuminated next to HDD 3 and HDD 4. Same DOA behavior. Clicking noise and not recognized by computer either via USB or SATA.

Would those red error LED lights next to HDD 3 and 4 indicate that two of the four drives just died, possibly hosing the benefit of RAID 5?
 
RAID5 allows for 1 drive to die with no data loss. If 2 drives are dead you lost everything.
 
peppergomez said:
Here is the device that I have:

http://www.mediasonic.ca/files/1383763501_0.pdf

In case these details are useful:

I am using Seagate Barracuda 3TB drives, model ST3000DM001

12TB total, but I think 9TB was accessible for storage.

I re-seated the drives, and booted the RAID up. Now there are two red Error LED lights illuminated next to HDD 3 and HDD 4. Same DOA behavior. Clicking noise and not recognized by computer either via USB or SATA.

Would those red error LED lights next to HDD 3 and 4 indicate that two of the four drives just died, possibly hosing the benefit of RAID 5?
Click to expand...

Doh! Those Seagate drives are well known clunkers. Some of the worst drives made in years:

https://www.backblaze.com/blog/3tb-hard-drive-failure/
 
mwroobel said:
Well, if it was 8TB advertised as you said above, and you only have 2TB drives then it is NOT R5. If it were R5, and one drive failed the array would be in degraded mode but still accessible. I just downloaded the manual, on the front of the panel near the top, there are LED's for 0, 0, 1, 3, 5 & 10. Which is illuminated? Next, just below that are the fan lights, and then below that are the disk lights. Which are illuminated, and for each are they blue, purple or red? Next, there is an LED to the left of each drive, that is the error light.. Which one is illuminated?
Click to expand...

Thanks for the follow up. 5 is illuminated, since it is RAID 5.

What do you mean by "degrade mode" and would the ProRAID give me any indication of this degraded state? I have never heard of this idea, so maybe it's possible that the ProRAID was already running in degraded mode, and that I didn't notice? I usually keep the front door on the enclosure, and have the enclosure away from where I can see it, so if any red error LED lights came on previously, I wasn't aware of it. I did notice dropouts when I first began using it, but those were due to shitty USB 3.0 behavior in Windows 10 (something I posted about on these forums, and learned about). I switched to SATA and these dropouts became far less frequent.

It would annoy me to realize that one drive had died/the RAID was running in degraded mode, without it or Windows giving me any feedback about that. I would have done something about it.

Image below:

lightbox


how the fuck do you post images? I posted a direct link to my flickr upload of the RAID. Jesus, I can't do anything right, it seems. Fuck. Here's a link, since this forum refusing my link, for some bullshit reason:

https://www.flickr.com/photos/87635527@N04/24319091370/in/photostream/lightbox/

(Irritation aside, how do you post images on this forum? I posted the direct link to my Flickr pic within the URL field specified. How can that possibly be incorrect?)
 
Degraded mode in the case of R5 would be 3 of the 4 drives operating, where the parity information on the remaining drives could be used to rebuild the array and keep it up and operating prior to the rebuild. Unfortunately, 2 of the 4 drives in your enclosure have failed per the information you are relaying and that is a death knell to R5. I am not familiar with the particular unit you are running, perhaps there is no software that will warn you on drive failures other than the LED failure indicators on the unit itself. In any case, the news is not good and I do not hold out any hope of recovery unless you can get valid data off the drives. How important is the data, and is it worth the many thousands of dollars it will most likely cost for professional array recovery?
 
I figured as much. I have crap luck with data. Over the course of several different systems over the years this is maybe the 6th or 7th such incident. First with a RAID, and the whole point of using RAID 5 was to avoid something like this.

Yes, the data is valuable but I can't afford the money it would take to recover it. I'm a photographer (www.madheiress.com) and there were several unedited shoots there. I manually back up the data every few months to another drive but several shoots were not backed up.

Shit luck, as always, when it comes to tech.

I appreciate the replies and assistance.
 
For what's it's worth, that array is likely doomed by the hard drives in it. Consider getting an additional backup.
 
Since this is a business, I am going to suggest that you dump the enclosure you have now. If you are not technically inclined, you might think about something like a Synology box (using NAS rated internal drives) with 1 or 2 external drives (or pairs of drives depending on your data requirements) for rotating backups. This will keep you up to the minute with notifications should you have any failures, will handle local backup and then also sign up with backblaze or a similar service for off site/DR backup.
 
mwroobel said:
Since this is a business, I am going to suggest that you dump the enclosure you have now. If you are not technically inclined, you might think about something like a Synology box (using NAS rated internal drives) with 1 or 2 external drives (or pairs of drives depending on your data requirements) for rotating backups. This will keep you up to the minute with notifications should you have any failures, will handle local backup and then also sign up with backblaze or a similar service for off site/DR backup.
Click to expand...

Idea...

Two Synology boxes. Primary box operating drives in Raid 1 pairs (something like a 1515+ or 1815+. Second box, nothing but single drive clones. All data backed up to "the cloud" (your choice of poison) that supports revision history.

If the Raid fails, replace the drive...1 volume affected only
If the rebuild fails or primary synology fails, copy off the clone(s) after new array(s) is built
If the building goes up in flames..cloud dude...cloud.

As for carbonite...they are not cheap and not sure what they really provide for small business that justifies their cost to match others. Their basic unlimited is "pretty slow". SOS backup, Crashplan, mozy, and slew of others exist. Take the time to make a spreadsheet that compares "key" apples-2-apples metrics. None of them are "perfect" though. The best cloud backup is one you own and maintain....take that worth a grain of salt though.
 
Last edited:
You must log in or register to reply here.
Back
Top