Hacker News Reader

Hacker News

Comment by ISL | Hacker News Reader

Full thread

ISL

The original "Drive Farming" link is at least as interesting as the new article:

http://blog.backblaze.com/2012/10/09/backblaze_drive_farming...

View on HN

And here's the discussion of that story:

https://news.ycombinator.com/item?id=4631027

I browsed the comments with two things in mind: was anyone uncomfortable with shucking and farming on this scale, and were the drives of equivalent quality?

Regarding the latter, user rsync of rsync.net claimed that "the drives inside these enclosures are the worst spec'd, highest failure rate drives." User budmang of Backblaze, however, said, "The drives were literally the same inside the external enclosures." User atYevP of Backblaze posted a similar comment.

High failure rates of external drives can also be attributed to the USB controllers - they're better than their IDE predecessors (which failed a lot, sometimes taking the drives with them - only low level utilities could revive them). Otherwise, I don't see how or why the hard drive manufacturers would bother to send the worst batches (which aren't that much worse) to external hard drive assemblers.

It could also be that they shut down more frequently. More stop-start cycles mean more wear on the moving parts of the disk.

Isn't that a feature of the USB enclosure? (i.e., not an issue once a drive has been "shucked")

I repair computers for friends and families and the majority of the time I get a broken external to look at, it's the enclosure that's at fault and the drive is absolutely fine.

I encourage people to buy their enclosures and drives separately - it's a little more expensive, but this way you can RMA just the enclosure if it fails. With an all-in-one, you either void your warranty to get the data back, or ship the whole unit back for a replacement.

That's an interesting point, since I have a Samsung hard drive I took OUT of an enclosure and out of curiosity I checked the warranty on it - it's a 3 year warranty that is still active and the web page said I could initiate a return.

So, technically, I could RMA the drive myself if it fails... or would they notice it and not process the return?

Yev here ;-) -> http://blog.backblaze.com/2013/11/12/how-long-do-disk-drives... We posted this earlier in the month. The drives inside the external drives that were shucked performed pretty well. We were guessing that they would fail more frequently but that just didn't turn out to be the case. We're currently collecting even more data and hope to release it soonishly :)

Interesting. I've never once found a standard drive model inside an external.

I mean, of course they are standard form factors, standard interfaces, etc- they are standard 3.5" hard drives- but they are always some odd model number I've never heard of with a 32kB cache, 1 month warranty, and 3600RPM. (slight exaggeration).

In my experience, most of the 3.5" enclosure drives have been off the shelf. My last batch of HP ones were all WD Greens. I have seen a few WD 2.5" externals with proprietary connections :/

Data recovery guy here...

Failure of external drives has mostly to do with their handling.

They get bumped around while runnning.

People trip over the cords and knock them over.

They stack them like pancakes and they overheat. They block the fans and cause overheating.

Many enclosures are designed to look stylish - but don't cool well.

Heads fly at nanometers over the surface - they dont tolerate bumps well.

The drives we recover from external enclosures appear to be the same as bare drives that we recover.

Reply on news.ycombinator.com