Jump to content
RESET Forums (homeservershow.com)

Another highly questionable post from BackBlaze


Drashna Jaelre
 Share

Recommended Posts

Hey guys, long time not posting for me but I happen to be listening to the HSS podcast and heard about the Backblaze article.I fully expected the 3T Seagate to be a bit higher than competitors but 40+%!!!.  I question this based totally on my own experience as I have as of this writing 18 of them running 24/7 for the past three years, and had as many as 30 for two years plus.  My failure rate combined over the past 3 years for all my drives is 2 hard failures and 3 additional drives that had correctable error counts that were climbing and I opted to replace them.  I am not saying the data is false, only that there must be some additional circumstances to see failure rates that high.  I can only comment on my experiences and if I saw a failure rates anywhere near 40% I would have put them all in the dumpster and went to a different brand along time ago.  I actually just bought two more about a month ago for some additional movie expansion.  Obviously all information must be taken seriously so if you are buy new drives don't buy these, but if Seagate had been receiving 40% RMA, across the board, they would have dumped that product and went out of business.  Just my two cents.

Link to comment
Share on other sites

From the data, it was mostly just the ST3000DM001 line, and I think one other line.  It definitely wasn't across the board, as well. As BackBlaze used a VERY limited set of disks.

 

That said, I've seen a lot of people with that line of disks failing....

Link to comment
Share on other sites

Yeah. It was mostly 2 models of Seagate HDD, the M001 in particular. I've said it before, but I'll say it again: Seagate needs a MUCH better way of identifying their models of drives. One of the reasons I buy WD is because it is SO easy to identify the various models of drive. For example, I don't buy Greenies. It's simple; if it's labelled Green, I won't buy it. If it's Red, I will. Seagate may think it's a bad idea to make drives that easy to identify because they might get lower sales of some models, but I would argue that that's a good thing. Seagate should weed out poor performers ASAP, in order to help the overall brand. The other thing is that, by not making it easy, they lose sales from people like me, who just can't be bothered traversing their model lines.

 

This is not an all out panacea, but it sure makes things simpler.

Link to comment
Share on other sites

The 3000DM001 is the only series I have used.  I still believe it might be skewed a bit.  Not defending them as they should be doing much better but just looking at a sanity check.


The 3000DM001 is the only series I have used.  I still believe it might be skewed a bit.  Not defending them as they should be doing much better but just looking at a sanity check.

Link to comment
Share on other sites

I was finally able to find this, this is the 2TB version of the same Seagate drive in question here:

http://www.amazon.com/review/R205L2P2CHJ02D/ref=cm_cr_pr_viewpnt#R205L2P2CHJ02D

 

This guy did a lot of research to determine factory and design issues. I have several of these Seagate 2TB drives, 0 failures.

I am sharing this, not to beat up Seagate, but to point out that drives with the same part number could be very different internal. This of course could mean that drive reliability could vary. This can be true of any vendor.

It would be interesting to see if there are different versions of the 3TB using the same part number.

Link to comment
Share on other sites

Thanks schoondoggy. That's quite a remarkable thing to find as a review on Amazon. It probably should be an actual article somewhere.

 

I didn't read every last word, but the gist of it conforms to much of how I feel about the BackBlaze blog.

 

Certainly it's a little disconcerting to realize that Seagate is now known to be giving physically different drives the same model numbers, only differentiating them by serial number. I have no idea if WD or others are doing the same thing, but it seems to be a somewhat deceptive practice.

 

And I completely agree, it could help explain how some people have very good experiences and other have dismal ones with what purport to be the same HDDs. It seems like it would be useful to have a web site listing the various makes, models, and variations of each.

Link to comment
Share on other sites

Another thought, back in the late 90's I was in a part of our company that sold a lot of drives to regional PC manufacturers. 10's of thousands of drives. These companies were manufacturers so our big concern was failure while in the factory. 4-5% was the norm, if failure rate approached 10% we would investigate. We would send drives back to the factory for analysis and we would inspect their manufacturing process to look for issues. The majority of the time the issues were do to drive handling. Drive failure testing would confirm it.

So what is missing from the Backblaze story(unless I missed it)? Backblaze should be getting requests from Seagate to test and determine failure cause. 

Why is there no mention in any of the stories about working with the vendor?

Makes me wonder,,,,,

Link to comment
Share on other sites

It would certainly help.

 

OTOH, the way BB buys drives, a couple thousand at a time from various sources, should mean that they are receiving drives that have been handled in a variety of ways. Hopefully, that makes handling amongst the different makes and models a wash.

 

Of course, in a sense, how a drive stands up to typical handling during shipping and installation is part and parcel of the whole reliability issue. It's a real-world experience for the drives, as opposed to a benchmark experience. It could be that some models of drive just stand up to rough handling much better than others.

Link to comment
Share on other sites

After reading their reports, I would never consider using them for hosting. This whole concept of shelling drives is crazy for a large company.

When you buy drives in quantity they should be shipped and stored in containers like this:

kZRtpa.jpg

 

Yes, some drives handle poor handling better than others, but all mishandling is detrimental to the life of all drives.

 

Purely my speculation, but there is a reason why Seagate is not working on this. I have to believe it is do to the way they buy drives or perhaps the compute pods they use. 

  • Like 1
Link to comment
Share on other sites

I agree, shelling drives is crazy for a large company, but I don't consider BackBlaze to be a large company. They were desperate for drives. They needed them and couldn't get them. It was what they had to do to survive.

 

BTW, I don't like that packaging for drives. If that styrofoam is the kind I think it is, it's pretty rigid, and it allows way too much shock to reach the drives. To me, that styrofoam package should be further buried in a cocoon of soft foam rubber chunks (the kind used in cushions and pillows), so that any sharp blow will be greatly softened.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...