Asustor 1002T

[Dref]

I have a Asustor NAS 1002 . Recently I have had two WD purples fail on this device within 3 months. (both less than 18 months old). I have been carrying out a lot of disk intensive operations (that's what purples are designed for) and I can't help thinking that this system just cannot cope with it. Anyone else have any Disk problems with the Asustor 1002T??

 

[RMerlin]

How are your disk temperatures looking like? This is the most common source of early failures.

If your model has a user-configurable fan, you might want to try setting a faster speed rather than leaving it on auto.

Also, if you bought both disks at the same time, it's possible they both came from a bad batch. Some server admins recommend getting HDDs from two different suppliers when configuring a mission critical RAID. That might be excessive, but still shows that such batch-related issues aren't that uncommon.

 

[avtella]

I've had a majority of my WD drives I bought or those that came in computers/Laptops etc fail, some of the first ones within weeks some in months and one or two lasted a year or so. This is my experience over 15 years. Maybe you are just as unlucky as I was. I doubt its the NAS.

Looks like WD's subsidiary the Hitachi HDD brand is the best.

 

[RMerlin]

One of the problems I have with Backblaze's data is they tend to get the cheapest drives they can find, regardless of the intended usage scenario by these drives. That kinda skews the results versus if you get a desktop drive and use it in a desktop environment. Their PODs probably put far more stress than a typical desktop (or even SMB NAS), cooling and vibrations might also be an important factor.

My own professional experience with WD has been much better than Backblaze's own. Aside for various Caviar Green having issues when used in anything BUT a USB enclosure for nothing more intensive than daily backups, the failure rate I've seen on WD HDDs has been way lower than Backblaze's own numbers. My experience with Seagates however do match theirs, including with Seagate's SAS products (which Backblaze do not use).

My sample size is probably around a bit over 100 HDDs over the past 5 years, ranging from desktop to rackmount servers. Out of these there are maybe a bit over a dozen Caviar Reds in service, not a single failure so far.

 

[avtella]

Damn 100 is a pretty good sample from you, I guess I was just that unlucky lol. Seagate used to be good at one time. The consulting firm my father worked for, sent him to Seagate (over a decade ago) and at the time IBM consultants apparently convinced Seagate to cutdown backend testing as they figured the increase in failure rates and replacement costs would be offset by the money and time saved doing the testing. Low and behold you get what you have now. Interestingly enough IBM's former HDD division which is the Hitachi sub brand now owned by WD, actually has a pretty good reliability rate compared to its peers.

 

[RMerlin]

Interestingly enough IBM's former HDD division which is the Hitachi sub brand now owned by WD, actually has a pretty good reliability rate compared to its peers.

Aside for their infamous Deathstar (Deskstar GXP75 - I had two of these die on me at home within a period of maybe 1 year - one died within three days), IBM's HDDs were among the best at the time. Glad that the quality was maintained despite going through two new owners since then.

My most problematic WDs were Caviar Green (in general) and their VelociRaptor (mine would die within 12-18 months of use). Their Blue, Black and Red lines have been very reliable for me.

Seagate's most problematic models were their infamous 7200.11 models (and 7200.12 but to a lesser extent). Also disappointed in the failure rate of their SAS products that I used in a datacenter - not catastrophic, but certainly higher than I would have expected out of enterprise products. I have more limited experience with their newer products, having given up on them after the 7200.11 issues.

Many manufacturers also suffered higher failure rate around the Thailand floods, either due to outsourcing to less quality production centers, or potential contamination of their installation. Backblaze do noted something about that in a past blog.

 

[Antonios]

I use this NAS model, one drive in it only, actually a drive I already have used for a number of years so I trust it enough to put it in there on its own.

Drive is extremely quiet and never heats up regardless of load, usually stays within 10 degrees above ambient, even in very hot weather.

If I had a second drive in there, I think it would add maybe 5 degrees overall due to more restricted airflow between drives. If it would go any higher than that I would take the cover off and keep it permanently off, adding some support beneath the casing for balance as the cover splits the bottom part and unit gets unbalanced when cover is missing.

In case you are interested, drive is a Toshiba DT01ABA200 of the 2 1-TB platter variety (not the HGST relabeled stock of the 4-platter HDS5C3020BLE630 which I believe it must run hotter and louder too).

Surely I had many drives in NASes and servers run much much hotter than that, most of them were SCSI 10K HGSTs and couldn't be touched with bare hand for more than a couple seconds, each drive had a dedicated fan on it too running full speed, the thing is no such drive failed on me before its time, that would be more than 5 years running non-stop ever.

But that was at my workplace at the time, in my privately owned NAS I would never like or allow this to happen, I like it cool and slow and this is good enough and feels safe enough for me.

Just my 2c.

 

[avtella]

I have the same 7200RPM Toshiba drives, Netgear sent them to all the RN104 test units. I'm using them in the 524X they sent now. Pretty reliable drives in my experience.

 

[sfx2000]

Seagate's most problematic models were their infamous 7200.11 models (and 7200.12 but to a lesser extent). Also disappointed in the failure rate of their SAS products that I used in a datacenter - not catastrophic, but certainly higher than I would have expected out of enterprise products. I have more limited experience with their newer products, having given up on them after the 7200.11 issues.

Many manufacturers also suffered higher failure rate around the Thailand floods, either due to outsourcing to less quality production centers, or potential contamination of their installation. Backblaze do noted something about that in a past blog.

Seagate had a bad run on 3GB drives a while back... and I had a lot of them in the data centers. Luck would have it that the failures where somewhat predictable, so we could be proactive and start rolling out older drives... but with lots of servers, one can get ahead of things - for those on the home/small business front, the Seagate 3TB failures were sudden and brutal.

Not that much different that the IBM deathstar's back in the day.

That being said - with most spinning rust - they're all about the same - and SSD's are improving in the enterprise space - I'm a bit concerned about consumer drives with the pricing pressure downwards...