4

I don't seem to have very good luck with hard disk drives these days; 3 side-by-side (non-RAID) drives on my machine, replaced 5 drives in the last 4 years. My problem of the moment is that my largest drive (3000GB) seems to now be having some pretty serious read issues.

At best, accessing it comes up with erratic read/write speeds, but for the most part, I can’t even really access the data for anything meaningful and testing it seems to fail out after a bit. The SMART test on it lists two warnings under “Reallocated Sector Count” and “Current Pending Sector.”

The drive itself is a Seagate Barracuda bought in 2013. Installed at the same time was a SSD which seems to be functioning just fine (I run performance and health tests on it every once and a while as it is my primary drive). The machine is a custom build with 1000w power supply and Cooler Master case. The machine and drives stay relatively cool and have good ventilation.

Is there anything I can do to save my drive, or am I looking down the barrel of yet another hard disk drive replacement?

1

2

  • 2
    Looks like it's time for a new drive. – Karan May 25 '15 at 22:25
  • @Karan Dang, that's what I was afraid of :( – LoveAndCoding May 25 '15 at 22:29
  • I posted an answer with my generic feelings on this situation, but as @BiTinerary states I would recommend editing your question to add details on the exact make/model of the drives themselves as well as details on the system setup—seems custom, correct?—as well as what kind of usage this system gets. A server needing drive replacements is one thing; that is expected behavior based on usage. If this is a basic—albeit souped up—desktop system that is failing like this, that is something else entirely. – JakeGould May 25 '15 at 22:40
  • Seems seagate has been making some shi*t drives lately. According to their specs, this drive is only rated for 2400 power on hours ( wtf? ) and yours has 7151. Combined with all of the bad sectors, looks like the drive is shot. I guess the lesson here is don't buy seagate. – psusi May 26 '15 at 0:56
  • Spinrite from grc might buy you some time, and recovery lost data, but it is not full fix. The long term fix is new hard drive. – cybernard May 26 '15 at 3:08
2

What specific kind of drive/model is this? Are your previous drives the same type of HDD? I assume yes if your running a RAID (what you meant by 3 parallel drives?) Regardless, this seems to be a reoccurring issue and simply "replacing the drive" isn't going to help. There's clearly an underlying issue that isn't being addressed, drives shouldn't be failing this frequently.

Perhaps it's an issue with the specific brand of drive. For instance Western Digital's green drives tend to have more "erratic" usage than others, especially if your settings or the way you access them isn't necessarily in tandem with their intended operations.

If you require this many drives perhaps consider educating yourself in HDD health, the difference between "Reallocated Sector Count and Current Pending Sector", more importantly basic HDD diagnostics and repair tools.

Here's some stuff to get you started:

Hard vs. Soft Bad sectors (part of what your SMART is flagging): http://www.howtogeek.com/173463/bad-sectors-explained-why-hard-drives-get-bad-sectors-and-what-you-can-do-about-it/

Free/Purchasable HardDrive diagnostics and scans that will tell you more than the surface scans of your SMART. They also offer potential repairs based on the Hard vs. Soft sectors mentioned. HDAT2, MHDD, SpinRite, HDDRegen, Hiren's Boot CD, Ultimate Boot CD.

EDIT: Seagate (3TB) drives have been notorious for some of the least reliable drives.

enter image description here enter image description here

http://www.pcworld.com/article/2089464/three-year-27-000-drive-study-reveals-the-most-reliable-hard-drive-makers.html

  • I clarified. It's actually not a RAID, I just meant 3 drives in my machine. A SSD primary, and two HDD for regular storage use. Thanks for pointing me in the direction for the SMART info, that was definitely my next step. I'll read up on those as well. – LoveAndCoding May 25 '15 at 22:50
  • Whelp, from the looks of that, looks like I've been buying the wrong brand :( – LoveAndCoding May 25 '15 at 23:16
  • Yea... 1/2, 1/3 failing drives within a year is not very good odds at all. I would suggest HGST/Hitchi (as seen above) I have them installed on 3-4 personal computers and every one is still kicking strong after 4+ years of usage. Not to mention none of the clients (that I've witnessed) of the company I work for has experience drive failure after we've replaced them with HGST's. Perhaps you can still contact Seagate inquiring about warranties? With emphasis on your experience concerning their unreliable drives? Also, why not buy 3x1TB drives rather than 1x3TB? – BiTinerary May 25 '15 at 23:31
  • 1
    Yeah, I was looking at that and will probably go HGST this time around. They look to be very stable. And in terms of 3x1 vs 1x3, I do that mostly for space in my case. I like to space them out so they can run without the heat compounding. I may downgrade to a 2tb instead this time (my 3tb only has about 1tb full at the moment, and I still have ~700gb free on my 2tb) but we'll see. I have a NAS RAID for backup now too, so it may be a smarter choice. – LoveAndCoding May 25 '15 at 23:41
  • I see, I see. I recommend running even one of the free scans I mentioned since I've seen SMART diagnostics give flat out FALSE readings. The scans do have a potential of repair and all it takes is a CD/USB and an overnight scan. Well worth it considering the time, experience and money you may still end up spending. Most importantly, these drives may no longer offer peace of mind regarding a backup but if the HDD scans fix some sectors and are still operational, why not keep them as "worst case scenario" or off-site backups? I have 2-3 drives like this that were "out of commission" years ago. – BiTinerary May 26 '15 at 0:15
1

First you say this:

I don't seem to have very good luck with hard disk drives these days; 3 parallel drives on my machine, replaced 5 drives in the last 4 years.

Then you say this:

Is there anything I can do to save my drive, or am I looking down the barrel of yet another hard disk drive replacement?

The bad and simplest news about this? You are definitely looking down the barrel at yet another hard drive replacement.

That said, it seems like the drive failure is a symptom of a larger problem since replacing 5 drives in the last 4 years is a tad unusual for everyday usage for a desktop system. If it’s a server running 24/7/365, then yes, drives will fail on a regular basis. But if this is a basic—albeit souped up—desktop system then that is something else entirely.

Are you sure your overall setup is working well? A faulty power connection or even data cable can contribute to conditions that would put additional unwarranted stress on the drives. Heck, does the case have decent cooling? Keeping three drives in one box with bad ventilation can cause overheating which—again—can put unwarranted stress on the drives.

  • Yeah, it's been pretty bad. I lost two and a video card to a faulty power supply a while back, which I replaced with a 1000w. The machine is a custom build, heavy use desktop. I used to run it 24/7 but haven't for several years now. I'll open my case up and see if there might be an issue with the wiring or anything. Things looked fine when I had a HDD failure a month or so back, but clearly two in a row indicates that it may not be. Thanks for the tips. – LoveAndCoding May 25 '15 at 22:53
  • Absolutely correct. Needs a little more info as to the setup. Completely valid point in regards to cooling and temperature. if you have 3 drives operating full blast, a standard desktop setup isn't going to be equipped with adequate cooling without some additions or modifications. Frequent heating/cooling is a MAJOR contributor of drive failure, next only to frequent spin ups/spin downs. Additionally, 3TB drives tend to have the highest RPM's out of any other drive which also causes them to have smaller life spans. If all of these things are true you have a brewing pot for drive failure. – BiTinerary May 25 '15 at 22:54
  • 1
    @BiTinerary, no, 7200 rpm is the "standard" speed these days ( well, for the last 10 years nearly now ) for desktop marketed hard drives, regardless of capacity. Drives marketed for backup, near line, nas, or low power ( like WD green/red/purple ) operate slower ( 5400-5900 rpm ). You can tell these drives because the manufacturers avoid specifying the rpm. Also drives use less than 10 watts, so even with 3 of them, that is still far less heat than you are going to get out of a busy cpu or gpu -- quite easily cooled unless they are all crammed tight with lots of cables blocking airflow. – psusi May 26 '15 at 0:51
  • “Yeah, it's been pretty bad. I lost two and a video card to a faulty power supply a while back.” If that is the case then something about your overall setup is definitely screwed up. I recommend spending some time—perhaps—rebuilding your system to really get this issue resolved. Otherwise you’ll just be buying enough new parts to build a second new system anyway if you think about it. – JakeGould May 26 '15 at 5:54
  • @JakeGould After that power supply failure, I replaced most everything in my machine and rebuilt it. Pretty much the only things I didn't replace were my case and a BD drive. Cables, PSU, MB, and most of the rest got swapped out at that point, including the two new drives (one of which is the one that is currently failing). When I swap in the new drive, I'll definitely be double-checking the whole build to make sure everything is connected properly, seated right, and is getting good ventilation. Thanks for the recommendation. – LoveAndCoding May 26 '15 at 17:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.