Hard Drive Failure: Need Advice

[jackflerj]

Surprised your 500GB drive went also.....

There are lots of complaints on tech boards about SATA drives >1TB and larger having high failure rates regardless of what brand you have. Drives that used to come with 3-5 yr warranties are now only giving you a year or two! This is a good indication that overall all newer large SATA drives are having quality issues. The old IDE drives by comparison were built like tanks. I never had an IDE drive fail.

For example, my 3 yr old Seagate 1TB SATA drive recently failed. Luckily it lasted about a month before totally crashing, which allowed all important files to be saved. W7 and Linux was dual booted on that 1TB drive. Linux first reported there were HHD issues and gave a full comprehensive report like you would get from a PC repair shop, for FREE of course!.... It reported issues were on the W7 side. Linux side still ran fine. Then a couple days later Dell on booting up, reported HDD problems. For giggles I booted up W7 which ran OK for a couple more days before M$ said there were problems. Then all M$ said was 'drive failure is imminent'! Backup & replace ASAP. A couple days later W7 refused to boot but Linux still ran fine allowing anything important to be saved on BOTH sides of that failing HDD.

Honestly I've only had a single problem with sata drives 1TB and over (I've purchased eight over the last few years), a 1.5TB Seagate Barracuda Green Drive. At the time green drives were new and unfortunately I didn't research the tech beforehand. Live and learn.

 

[WoodPeckr]

Honestly I've only had a single problem with sata drives 1TB and over ...

That OEM 1TB sata Seagate drive was the first drive failure I ever had in 14 years. It lasted only 3 yrs. Then when doing research for a replacement, was when I discovered lots of complaints on tech boards about Drives 1TB and larger having higher failure rates than in the past. Anywho, got a 2TB Seagate sata HDD for replacement and hope it lasts longer than 3 yrs.....:eyebrows:

 

[Radio_Shack]

Sorry that I didn't read all the responses here but I have had success with sick drives by putting them in an enclosure and mounting them under ubuntu or other linux bistro and copying the data to another disk using linux cp command. For some reason linux has an uncanny ability to read things that windows can't. Anyways thats my $0.02 $CAN worth. Good luck if you have not already got something to work. Cheers ..RS..

 

[jackflerj]

That OEM 1TB sata Seagate drive was the first drive failure I ever had in 14 years. It lasted only 3 yrs. Then when doing research for a replacement, was when I discovered lots of complaints on tech boards about Drives 1TB and larger having higher failure rates than in the past. Anywho, got a 2TB Seagate sata HDD for replacement and hope it lasts longer than 3 yrs.....:eyebrows:

Same here. So far so good on both drives:

Western Digital WD Black WD2002FAEX 2TB 7200 RPM 64MB Cache SATA
Seagate Barracuda ST2000DM001 2TB 7200 RPM 64MB

The others are in machines I built for other people. Over 3+ years on all without complaints.

 

[thewheelman]

UPDATE

I shut my computer down, plugged and unplugged the hard drive in question, rebooted, and lucked out as the drive was detected in BIOS and by Explorer. I immediately began copying over data and got almost all of it but there's a little bit left that won't copy; the file transfer hangs whenever I try to get the remaining files.

So this tells me there must be some kind of corruption on the drive that's causing the problem. The clicking sound shows up once it hangs and then never stops, like it keeps trying to read the disk but can't.

Any idea what this could be? Bad sectors? Would running a scandisk help?

Did you try using the SMART tools to read the drive's fault report?

You can try a boot disk tools suite like Hiren's, and use it's Hard Drive SMART tools, or try here http://en.wikipedia.org/wiki/Comparison_of_S.M.A.R.T._tools

 

[WoodPeckr]

Did you try using the SMART tools to read the drive's fault report?

This comes standard on Ubuntu and gives you a very comprehensive report on the health of your drives and system.....
It detected and gave a highly detailed report on issues with my failing 1TB HDD days before Dell or Windows 7 even noticed the drive had problems.

 

[The Options Menu]

This comes standard on Ubuntu and gives you a very comprehensive report on the health of your drives and system.....
It detected and gave a highly detailed report on issues with my failing 1TB HDD days before Dell or Windows 7 even noticed the drive had problems.

Though there are free SMART reporting tools for pretty much every OS. It also never hurts to use the SMART tools on OEM drives from cheap places. Some of these places will try to sell a second hand drive as OEM. The SMART tools will tell you if a drive is new or not.

The lessons here are:
-Data matters, software doesn't.
-Always have a reliable, preferably local (non-internet / cloud), and preferably automated backup system.
-Replace drives on a schedule, not at the point of failure. (I'd suggest every couple of years.)
-If you have the option, don't format the replaced drive but securely store it off site. (That's the 'my place burnt down' failsafe.) Failing that, then consider and internet / cloud backup option, but remember that puts your data at the mercy of a company with a funky 117 page agreement, that can go bankrupt, or have legal troubles, and that you're relying on one or more intermediate networks to get to that data.
-If you can figure it out, use a mirrored RAID. (Trivial under Linux.)
-If you can, set up your permissions so you have to log in as a special user to destroy your data.
-Check what the dive has to say about itself via SMART, but automate it in a manner that you can't miss if you can.
-Don't use spinning disks in bouncy / vibrating places. I don't care if the device has 'jolt protection', just don't.

I'm a data nut since I lost several hundred hours of coding, along with the rest of my files, in the mid-90s. Never again. That song that girlfriend wanted me to download in 2003. Still got it. Still neatly filed. Never delete, never lose data, never surrender.

 

[The Options Menu]

That's a pretty decent explanation, although some might suggest that there are differences not only between Operating Systems (such as Windows and Linux) but also File Systems (such as NTFS or EXT3) which apply to how efficient the disk operations are. Some file systems will actually try to split a large file over a number of "nearby" locations on the disk, so that reading it back doesn't add a lot of latency and defragmentation is unnecessary.

Basically, if it's anything but a Fat32 (or earlier Fat) filesystem on a spinning disk, don't even consider defragmenting. That would be Windows prior to Windows 2000 (not Me), or some newer Windows (usually XP) boxes with very weak hardware that have a hard disk and not a SSD. You still find Fat32 on most SD cards and USB keys, but they usually do some funky things when it comes to file allocation, so don't worry about those. As was said, most 'modern' filesystems do a good job of not barfing a file across half the disk platter so banish word defrag to history.

Any other filesystem that's not Fat32 and still needs defragmentation probably belongs in a computing museum. (If you can still find an XP netbook, many of those used Fat32.)

 

[IM469]

I was running into similar problems but while the 2 TB drives are becoming the norm, they are becoming relatively cheap. I use a portable USB 2 TB to backup my fixed drive ~ 3 weeks which I use as a program source for my living room TV. My critical financial files and work files are backed up to a USB memory stick. These are cheap (16/32 GB ~ $15-$25) and are useful for catastrophic computer failures.

 

[WoodPeckr]

I was running into similar problems but while the 2 TB drives are becoming the norm, they are becoming relatively cheap.

Indeed they are cheap!
Decided to get a 2 TB drive for 69 bucks because it was cheaper than getting a 1 TB drive....

 

[enyaw]

aren't the larger sized drives more prone to give in on you. Aka higher failure rate because of the # of platters etc? Not talking about ssd's

 

[WoodPeckr]

That is the big complaint on many tech boards lately.
Guys are complaining that all SATA HDDs 1TB and larger are not lasting as long as the smaller HDDs used to last.

 

[enyaw]

I think the 2t drives have a place. They are ideal for backups and such because you wouldn't be using them every day. The wear and tear on them would be reduced, let's say you turned on your backup drive once a week. And the pricing is ideal. My 2 cents.

 

[The Options Menu]

I think the 2t drives have a place. They are ideal for backups and such because you wouldn't be using them every day. The wear and tear on them would be reduced, let's say you turned on your backup drive once a week. And the pricing is ideal. My 2 cents.

Unless you measure your data in TB. I do. I use a fully mirrored RAID with 2 multi-TB disks, with another multi-TB backup through the wall in the next room. Always assume any disk will die at any second, figure out your tolerance for loss, and plan accordingly.

In my case that means 2 identical new disks every couple of years, with one old disk going into the backup, and the other old disk going to my parents house. (With the 'old old' disk being given away) Roughly. Depending on what I'm up to. I'm quite content to let the SSD I have my OS on die of old age-- The OS doesn't matter, the data does.

The wrong time to try anything is after your disk craps it's pants, or you nuke your own stuff. (I'm paranoid enough that I have to log on as a special user to nuke my data.)

 

[Rockslinger]

In my case that means 2 identical new disks every couple of years, with one old disk going into the backup, and the other old disk going to my parents house.

Shoot, what kind of stuff do you have on your multiple harddrives? A cure for cancer? Just kidding!

 

[enyaw]

I agree with what the options menu is saying. It all depends what's going on the drives and of what value it is.. If you do alot of work with computers, meaning it's your job. Whatever the field might be. The data is the most important thing by far, time is money. And some of that work can't be replaced. BACK UPS are a must, some people might even get fired if they lose data. Multiple backups are a good way to go too. Be proactive not reactive

 

[The Options Menu]

Shoot, what kind of stuff do you have on your multiple harddrives? A cure for cancer? Just kidding!

Well, the two drives are mirrored and encrypted, so it only looks like one drive, then one encrypted backup. I have a 'never delete' policy, unless it really is that bad (in the Gigli sense). So I've accumulated a lot of music and video since the 90s. I actually don't mind losing that-- I also have thousands of photos I took, and thousands man hours of code that is stuff I use as utility programs, and stuff I am, or have been, toying with that has never been released on any public servers. That's ignoring everything I've ever written since high school, or anything else I've ever worked on since the 90s.

If that went 'poof', what would you do? The last time I lost my data (in the 90s) it was like losing a friend... Almost half a lifetime later what would that be like? Not better.