Three disk failurs within 15 months on same channel

How did they fail? reallocated sectors or something else?

Here's my theory & I'm interested to get feedback from others:

I think that since the floods in Thailand, HD quality has gone way down the tubes. Granted, my IT support biz has been booming, so maybe it's just because I see so many more computers, but I've seen a lot of dead HDs and new disks with reallocated sectors; most of them Seagate.

I don't know if the manufacturers were in such a big rush to get back online after their factories flooded that QC slipped, maybe crap got into the manufacturing equipment, I don't know, but that's my thinking. I have seen a LOT of seagates fail too.

Anyway, here's a recent experience I had:

ReadyNAS Ultra 6 reports a drive with SMART errors, and it eventually dies. I RMA it to Seagate, no big deal.

"New" (refurb) drive arrives. I install it into the ReadyNAS, same channel - 3.

Volume syncs up, within a day or two I start getting alerts of SMART errors on that exact same disk.

"No way" I think to myself.

Within 2-3 days, the disk is dead.

I RMA it to Seagate. Another refurb disk arrives.

Throw it into the same channel, it is DOA - the drive starts beeping as soon as I install it. NO. F-ING WAY. (Granted, it arrived in a really beat-up box)

Try it in each different drive channel, no dice.

The THIRD drive arrives, and FINALLY so far it seems to be good.

Basically, I've had the same experience as you. I blame the drives. I very firmly believe that QC has gone downhill, especially with Seagate. At least their RMA process is fairly streamlined & painless.

HHDs are voluminous nowadays and pretty cheap. That's what the competition between HDD brands gave us.
But reliability of cheap disks is not the same as it was some years ago, when 1 GB disk was more expensive than 1 TB disk is now. If you need reliability you can try a bit more expensive line of HDDs. That's my penny to the thread.

Maybe we can give the OP a chance to respond?

StephenB wrote:
How did they fail? reallocated sectors or something else?

Yes, reallocated sectors...

During these ~4 years I've had the NV+ the two first years it worked flawless. Then 2 years ago I had to disks starting to go bad. One in channel two and one in channel three. Changed both. After that I've had another two disk failures, due to reallocated sectors, as stated above all. Refurbished disks I assume.

49C does seem pretty hot. My drives are running 32-41C at the moment (various NAS, including an NV+ v1) - though they are mostly WDC Red and Green drives (with a couple of Seagate ST32000542AS and one ST31500341AS thrown in). Maybe other NV+ owners with similar drives can also share their drive temps. There may also be something not quite right with your fans. Have you ever recalibrated?

The main study I know of for temperature is here: http://static.googleusercontent.com/ext ... ilures.pdf
It does show a significant increase in failures for 3-4 year old drives running at 40-50C, but no significant temperature effects for younger drives. (see figure 5). One challenge with re-certified drives is that you have no idea how old they are. One thing to keep in mind is that this paper was published in 2007, so those drives are 10-year old technology at this point.

I installed a few re-certified drives in my Pro when I was first setting it up to save on cost. My experience was similar - most of them failed in 18-24 months. Now I either deploy the RMA returns somewhere else (a desktop PC or media player), or keep them as emergency spares.

I am thinking that you should perhaps start replacing these drives with cooler models as they fail. Either the WDC Red or perhaps the new Seagate STx000VN000 NAS lines would be reasonable choices. I've been deploying the 3 TB Red drives for about 9 months now, so far my results have been good. Of course that's not long enough to gauge long-term reliability.

Hmmm, Netgear seams to say up to 60C is normal. The fan has been recalibrated and I did it right now. No difference, same temp. I don't know if its speed says something, 2143RPM!?

But again, it's only channel three where the disks fails and its on that channel the temp is a couple of degrees hotter which I think i strange. Also with a new replacement disk inserted.

Well, it's Seagate that specs the drive operation conditions.

The ST31000528AS operating range is spec'ed 0-60C (http://www.seagate.com/staticfiles/supp ... 36864b.pdf) . Though I don't think that means that Seagate is warranting that sustained high temperature operation won't shorten drive life.

Ok, thanks for your response. We'll see if someone else comes by who has the same strange problem as I...

I've had something similar in my NV+, albeit on a much shorter timescale. When I first bought it, I put a pair of 2TB WD drives in bays 1 & 2 and left the other 2 empty. After about 3 weeks, one of the drives started generating SMART errors and failed almost immediately. Instead of RMA'ing it to WD, I returned it to Amazon for a replacement with a new device (I don't trust the reliability of refurbished hardware as all the junk migrates to the spares loop). The second drive displayed almost the same behaviour, as did the third - each arrived "as new", well packaged and in a undamaged static bag. The fourth drive worked perfectly and hasn't missed a beat for the last 2.5 years.

These were "early life" failures, and I strongly suspect a batch problem because of the behaviour I saw and the way I sourced replacements. In the OP's situation, I think the ~9 month gap between failures is a bit long to be regarded as "early life", but the temperature difference in the bay doesn't seem significant and I think it's more likely that you've been unlucky with the replacement drives than there being a problem with your NAS hardware....