I had 2 hard drives fail today

[Wag]

Hitachi!!!!!!

Both had the same SMART errors too. Has this ever happened to anyone else before?

 

[Smoblikat]

Been using a 2TB hitachi as my primary storage for a few years now, still have all sectors green and no errors.

 

[C1]

Ditto. Have similarly been using three of them with no issues, but they are older (ie, first generation) ones (eg, 2010 manufacture).

Now days Im careful about purchasing drives (eg, before the flood vs after) and prefer older pre-flood units (made in Malaysia or Thailand preferably pre 2011).

Reliability is more important than speed.

 

[KingFatty]

Has anyone ever done the math to know what the expected failure rate is for each drive to fail after X hours of time, and then calculate the odds that both drives will fail at exactly the same time like that? Go buy a lottery ticket man.

 

[Aikouka]

Did you buy both drives at the same time from the same place? I've always heard that it's recommended to avoid doing that if you need them to work in tandem (RAID arrays) because if one fails, the other statistically has a higher chance of failing as well.

 

[Wag]

They were of two different sizes, one 2TB, one 3TB. I think I bought them both from Newegg (if that matters). They weren't in a RAID array, although they were in the same external Mediasonic drive box.

I was thinking it's possible the eSATA cable is bad because it's a long run (10ft), but there are other drives in the box which show up as OK, so that can't be it.

 

[SheHateMe]

Thats terrible. I have a 2TB Hitachi and now Im worried. What program do you use to test the HardDrive health?

 

[hhhd1]

2 drives from the same manufacturer, subjected to the same environment, and the same power and the same type data cable, could go bad at the same time if they were subjected to the same fault in any of the same factors mentioned above.

 

[Compman55]

500GB drives are still more statistically reliable. 2TB and 3TB are quite a bit worse overall from all companies.

How did they fail? Could it power related, or cooling related?

On the same day tells me a power surge or controller malfunction writing giberish to the drive.

 

[Wag]

First Win7 started popping up with the Drive Failure Imminent box, then I used a few different tools like CrystalDiskInfo to check SMART status. Both drives had bad Reallocated Sectors Count. I have a third, older 1TB drive which is giving me a warning on the same thing.

 

[Smoblikat]

First Win7 started popping up with the Drive Failure Imminent box, then I used a few different tools like CrystalDiskInfo to check SMART status. Both drives had bad Reallocated Sectors Count. I have a third, older 1TB drive which is giving me a warning on the same thing.

Sounds like you need to stop your box ASAP. And switch the remaining drives over to an internal setup if you need the data on them.

 

[C1]

It is quite reasonable that two drives from the same production lot (ie, manufactured same production line close together) could fail. This actually is quite classic. An improper condition with the assembly line or process can typically result in producing a string of drives (or units of anything else) with a similar quality defect. QA testing is typically lot sample testing too.

I would imaging this might be the situation with the OP if the unit serial numbers are close.

 

[tweakboy]

Wait let me clarify.

bad sectors can not be fixed by windows. It can only be fixed with a app called

HDD Regenerator ,, I payed for it 15 bucks. You run that and next time you do chkdsk you will see it says 0 bad seconds your percantage healthy will go up too.

 

[tweakboy]

put it before you sleep cuz it takes very long time you wake up its done.

 

[Cerb]

Wait let me clarify.

bad sectors can not be fixed by windows. It can only be fixed with a app called

HDD Regenerator ,, I payed for it 15 bucks. You run that and next time you do chkdsk you will see it says 0 bad seconds your percantage healthy will go up too.

...but if the sectors went bad, you're just getting a false sense of security. Of course Windows doesn't fix bad sectors. It marks them, and keeps a table of them as places not to write to. The hardware is what should handle them. It tries, and usually succeeds, keeping your DATA safe even when the drive is failing, by trying over and over again, then writing the data somewhere else.

Some programs like that can help repair partition and FS problems, when a bad sector unexpectedly eats data, but the best fix is to take an increase in bad sectors as an indicator that the drive should be replaced, and/or that something else in the system is faulty (such as PSU or RAM, or around here, a too-high OC).

If you must recover the data, and dd doesn't cut it, then by all means, spend the money. But, the rest is just marketing, often deceptive, 99% of the time. If a re-write and remarking "fix" the problem sectors now, whatever caused them in the first place has still not been fixed.

 

[Gigantopithecus]

Two drives dying on the same day is likely to be due to something other than the drives.

 

[Wag]

They are both still under warranty so I will send them back and get new drives.

 

[Emulex]

it's not uncommon for sas expanders to freak out when drives go out, or for a cheap expander to nuke another drive when one fails. I've even seen raid controller cache boards nuke 1 or 2 drives while it takes itself out.

Thank god for raid-10 - you can lose a cable(connector,logic to such cable), the chassis (most decent sas expanders are dual ported with split logic and alternate your drives) etc.

I use 12 A7K2000 for D2D so they get extremely heavy writes every day since I do only full bare backups (no incremental/deltas). Solid drives.

One of the main benefits of using HP drives in HP chassis with HP smart array controller is that the drives are tweaked on both the raid controller, and drive firmware to be a little more reliable. Oddly they blew out the 2TB @ $200 a pop (with sleds), and 3TB at $150 a pop in the past few months, cheaper than buying generic drives and fitting sleds and having non-specific firmware.

DL320s with 12 146gb 15K SAS 3.5" - throw sas drives in garbage, add 12 new drives, good to go $299 for base server with 12 146gb drives. Probably could sell those drives for $299 lol.

 

[NXIL]

Two drives dying on the same day is likely to be due to something other than the drives.

I am with The Giant on this one.

I just had the old nForce4 SATA controller slowly die on me; took a few weeks--was getting hard drive errors, but very intermittent. Then, BIOS would not find hard drive on boot--I thought drive was going bad.

Tried with a new one: no go.

Old "failing" drive on different motherboard: worked fine.

So: how old is the motherboard or controller?

If not that: dirty power? Bad memory leading to bad writes on drive?

Two going bad at once, on the same day: hmmm, buy that lotto ticket!

 

[Wag]

Motherboard is practically brand new, top of the line Gigabyte Sniper. I doubt it's a PS either. I actually just popped the second drive in from a separate eSATA housing, it came up with a bad Reallocated Sectors Count as soon as I checked it. It could have been coincidence and been bad all along, I just never checked it before.

The third drive I am having a "Warning" on a SMART check for Reallocated Sectors Count is an internal drive in my tower (not even in my drive box). It's not coming up bad as of yet.

Is it normal to have this many SMART errors, Reallocated Sectors Count in particular, or am I just unlucky?

 

[Cerb]

Very unlucky, or something is faulty. It is common indicator on drives that are on their way out, but it should not be happening on several different drives at once.

 

[jjsbasmt]

I would almost think of a heat or power issue of some kind. Do you run them continuously or do you power them on/off many times a day? I'm not asking about normal spin down, rather cold startups.

 

[Wag]

Nope, don't run them continuously. The only thing I do occasionally with my drive box is change from the eSATA to USB port.

 

[ohforfs]

Been using a 2TB hitachi as my primary storage for a few years now, still have all sectors green and no errors.

Same. And I usually use Samsung. My failures are usually on WD or Seagate.

 

[ohforfs]

Two drives dying on the same day is likely to be due to something other than the drives.

Yup, definitely.