• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Just lost my first hard drive

EnderW

[H]F Junkie
2FA
Joined
Sep 25, 2003
Messages
11,459
I came home from school today and after I was using my computer for a few minutes, I noticed an unusual noise, like a "chirp" coming from it.

After removing the side panel, I tracked it down to the hard drive cage containing my four new WD3200JD drives setup in RAID5. The noise was not constant, but it was consistent, and I could imagine the platters making such a noise if they were rubbing against something while spiining.

So, I decided to take the drive cage out and disconnect each drive until I could determine which one was making the noise.

But first, I decided to take a nap, because I was tired. I shut the computer down, because the noise seemed a bit louder as I lay there trying to sleep.

I just woke up and turned on the system to be greeted by "ARRAY CRITICAL" or something to that effect. The good thing is the noise is gone, so hopefully it was just one bad drive.
I'm not too worried about my data, since I copied it all from my old array composed of 4 x WD2500JD drives and a LSI controller. I've disconnected them all now, but I should be able to hook them back up and use that array if need be.
And most stuff is also backed up on my 300GB Seagate drive except for a bunch of stuff I've downloaded from usenet, but it's not hard to replace.

So, what do I do now? Should I unhook the array until the replacement drive comes?
How can I check the other drives to see if they're about to fail? I've only had them hooked up for about 2 days.
I noticed I can access all the data on the array, how is this possible?
 
The array is running in degraded mode. With RAID-5, the array and data remain functional and available as long as all but one of the array's members are online. Performance will take a dump, as the data must be computed from the rotating parity disk in addition to being read off the remaining members, and you're gambling with failure the longer you let the drives spin without fault tolerance, but it will still work. Once you get your replacement drive in, leave the array alone while it rebuilds, that will take a few hours.
 
DougLite said:
The array is running in degraded mode. With RAID-5, the array and data remain functional and available as long as all but one of the array's members are online. Performance will take a dump, as the data must be computed from the rotating parity disk in addition to being read off the remaining members, and you're gambling with failure the longer you let the drives spin without fault tolerance, but it will still work. Once you get your replacement drive in, leave the array alone while it rebuilds, that will take a few hours.

I was planning to just order a repacement from ZZF since they have free 2 day shipping which should be a lot faster than RMAing this one to Monarch. This is the last time I buy a drive from anyone but ZZF. Only reason I bought them at Monarch is because they were on sale for $140 a couple of months ago, which was a good price then, but now it's a $1 more than ZZF. :( :mad:

They weren't packaged as well as ZZF, they were taped together in pair and wrapped in bubble wrap.

Anything I can do to test the other drives once I restore the array?
 
On the RAID controller? I doubt the manufacturer diagnostics will work. You'll have to rely on SMART monitoring output from your controller, if any.

Have you considered RMA'ing direct to WD? Advance Replacment will have you back in action in 72 hours.
 
DougLite said:
On the RAID controller? I doubt the manufacturer diagnostics will work. You'll have to rely on SMART monitoring output from your controller, if any.

Have you considered RMA'ing direct to WD? Advance Replacment will have you back in action in 72 hours.
The controller doesn't seem to have anything as far as testing the drives.

Good idea on RMAing direct to WD.

A question about my controller settings - under cache settings I have four options:
  • None
  • Read
  • Write Back
  • Read + Write Back
Right now I've got it set on Read + Write Back, and it seems to be fine, but I'm not sure what the difference is.
 
The cache settings should be of little impact in your application, as most of your transfers will be of files far larger than the cache size and the sequential thoughput of the array should be the limiting factor.

Read caching is for frequently accessed data on servers with large multi-user loads (for example, a corporate intranet homepage would probably spend a good deal of time in a controller's cache memory), while the write back cache is probably an extension of the lazy write capabilities of the disks, although I am not 100% sure here.

I am extremely disappointed that WD is going to have to eat another warrnaty claim because of poor HDD shipping. Crap like that drives up prices for everyone :mad:
 
DougLite said:
The cache settings should be of little impact in your application, as most of your transfers will be of files far larger than the cache size and the sequential thoughput of the array should be the limiting factor.

Read caching is for frequently accessed data on servers with large multi-user loads (for example, a corporate intranet homepage would probably spend a good deal of time in a controller's cache memory), while the write back cache is probably an extension of the lazy write capabilities of the disks, although I am not 100% sure here.
OK, thanks.


DougLite said:
I am extremely disappointed that WD is going to have to eat another warrnaty claim because of poor HDD shipping. Crap like that drives up prices for everyone :mad:
Yes, it makes me mad to, since it's easy too avoid. The packaging ZZF uses should be standard issue for hard drives.
I intend to spread the word as much as I can to only buy hard drives from ZZF.

I am somewhat curious as to how they package muliples drives, though. I wonder if they each have their own little foam "house" or not. I'd expect so, but if not, you could always just place seperate orders since they have free shipping.


On a side note, WD's RMA process has been very easy, and straightforward so far.
 
EnderW said:
OK, thanks.



Yes, it makes me mad to, since it's easy too avoid. The packaging ZZF uses should be standard issue for hard drives.
I intend to spread the word as much as I can to only buy hard drives from ZZF.

I am somewhat curious as to how they package muliples drives, though. I wonder if they each have their own little foam "house" or not. I'd expect so, but if not, you could always just place seperate orders since they have free shipping.


On a side note, WD's RMA process has been very easy, and straightforward so far.

In my case, yes, I bought 4 drives for a customer and each came in those wonderful little foam houses, very cool way to ship. Even the floppy came in one, blew me away.

I love the Egg, but ZZF owns HD shipping as far as I can see.

Doug is right, way too much money and data lost to something as stupid as bad packaging, it's very sad.
 
feigned said:
ZZF for hard drives you say?

How do they pack them? Pictures?



And of course they use antistatic wrapping too, this is just an old foam thing I saved.
 
^^ You will observe that the drives are not allowed to move in the packing, yet they are protected from being crushed or punctured while inside the box. I think it's time for a cheesy Tina Turner quote: ZZF is "Simply the best, better than all the rest" when it comes to HDD shipping.
 
You said you disconnected all the HDs? From the RAID controller? Make sure you put the cables back in in the correct order. I think we had a forum member who didnt number the drives to the ports and lost all his/her data.
 
doormat said:
You said you disconnected all the HDs? From the RAID controller? Make sure you put the cables back in in the correct order. I think we had a forum member who didnt number the drives to the ports and lost all his/her data.
I've disconnected the old drives from the old LSI controller.
I numbered each drive with a marker incase I need to hook them back up.
 
If your budget allows for it you might want to consider having a hot spare in the array so the card can automatically rebuild in the event of a drive failure.

On an unrelated note, are you the "EnderW" that is involed in python scripts for XBMC releases?
 
Ender said:
If your budget allows for it you might want to consider having a hot spare in the array so the card can automatically rebuild in the event of a drive failure.

On an unrelated note, are you the "EnderW" that is involed in python scripts for XBMC releases?
That's a good idea, but my controller is only 4 ports.
I do, however, have a seperate 300GB Seagate I use to back up the important stuff.
I'm also going to be creating DVD backups of everything irreplacable.

No, I'm not - common name apparently.
 
EnderW: What kind of RAID controller are you using for your WD3200JB RAID 5 array?

Just curious to know.
 
l337zax said:
EnderW: What kind of RAID controller are you using for your WD3200JB RAID 5 array?

Just curious to know.
Broadcom RAIDCore BC4452
 
EnderW said:



And of course they use antistatic wrapping too, this is just an old foam thing I saved.
That's pretty nice. We had a couple of SCSI drives come in like that (static wrapped of course) in a box. I was impressed.

If ZZF ships all disks this way that's where I'm getting them from in the future...

Sorry if I derailed the thread. :)
 
Update:

WD's Advance Replacement is slow. I requested an RMA on Aug 24, and they haven't even shipped the replacement yet - their site says to allow 3-5 business days.
:( :mad:
 
EnderW said:
Update:

WD's Advance Replacement is slow. I requested an RMA on Aug 24, and they haven't even shipped the replacement yet - their site says to allow 3-5 business days.
:( :mad:[/QUOTE

Did you call or email? One year ago, I had a 2 year old 120 GB drive starting clicking, I called to RMA, and they had a drive at my door in 2 days.
 
NACZ3 said:
EnderW said:
Update:

WD's Advance Replacement is slow. I requested an RMA on Aug 24, and they haven't even shipped the replacement yet - their site says to allow 3-5 business days.
:( :mad:[/QUOTE

Did you call or email? One year ago, I had a 2 year old 120 GB drive starting clicking, I called to RMA, and they had a drive at my door in 2 days.
e-mail
 
:mad: :mad: :mad: :mad:

Drive still hasn't shipped so I just called WD and they said it appeared the drive was out of stock, but he couldn't tell me for sure and would send a message to the RMA dept and call me tomorrow.
Dumbasses didn't bother to tell me it was OOS. E-mailed them twice with no response. Finally had to call them and then they can't give me a straight answer.

With shitty service like this, I'll more than likely look to Hitachi for my next storage drive(s).
I'm sure I'll still buy WD's new Raptor, just because I'm a hard drive whore, and I want the fastest thing available.

So, depending on what they say tomorrow, looks like I may have to order a drive from ZZF.
I may also go ahead and order a cold spare incase this happens again. I would put it in as a hot spare, but I don't have the room in my case and my controller card only supports 4 drives.
 
I had a drive die for me when I was creating the array. However, I think it was cause I barely tapped it during the creating of the array.

I've had no problems since, it's been up for 58 days, and there's a few more left until I power down and backup a few months of data to tape, and put the other stuff to one of the 400GB I keep a second spare copy on (powered off 99% of the time, only to update stuff).
 
They finally shipped my drive today.
I requested the RMA on Aug 24, so it only took them over two weeks - and that's for the advanced replacement.

I think I've decided to trade my 300GB Seagate I use for misc junk for another WD3200JD. Then, if I lose another drive, I can just clear off whatever is on it, and swap it in while I RMA the bad drive.
Hopefully I won't lose any more drives, though - the other 3 have been fine so far. Running this array at critical status has made me a little nervous, since it's the same as RAID0 until I get that drive back in there.
 
I have done a few WD advanced replacements via email and had them at my door in or before 3 days.
 
EnderW said:



And of course they use antistatic wrapping too, this is just an old foam thing I saved.


It's to bad they use the exact same foam house for laptop harddrives. Those thing bounce around in there like all hell.

I just got mine yesterday, hopefully it's gonna work fine after a trip in the foam walled room.
 
strossos said:
It's to bad they use the exact same foam house for laptop harddrives. Those thing bounce around in there like all hell.

I just got mine yesterday, hopefully it's gonna work fine after a trip in the foam walled room.
It should be fine, I'd think the foam would soften any bumps.
Plus it's a laptop drive - they're made to take a beating.
 
I have always had good luck with advanced rma with wd as well. They are better than seagate I have just recently learned, seagate charges you a 25 dollar fee to do advance rma.
New sata2 drive I just purchased was a hitachi, so far I am happy with the perfomance.

on a side note.. you always come across people wanting the fastest harddrives, where cost is not much of a factor.. why is it there are not more people that just go straight scsi to get the fastest out there?
 
$25 for advance RMA from Seagate?!?! Boo!! Hiss!!

As for not running SCSI, there are several factors that limit the adoption of SCSI by enthusiasts. The first is the small selection of drives that are less than $200, which basically limits you to 18 and 36GB 15K drives, and 73GB 10K drives. The 10K SCSI drives aren't significantly faster than WD740GD. The second is that you have to buy a SCSI HBA, while SATA is basically free on any worthy enthusiast mobo today. Next, you have to setup the SCSI bus and buy pretty pricey cables. Most of us don't want to worry about termination, setting IDs, etc. Another significant weakness of SCSI drives is their high heat and noise profiles.

Yes, Fujitsu MAU delivers the absolute fastest desktop storage performance in a hard drive that money can buy, but it will set you back a minimum of $200 to get 36GB, dramtically increase the complexity of your system, and also create a significant heat and noise problem. You can say what you want about WD740GD being a mediocre to poor enterprise drive, but it has conquered the power user roost, as small as it is. SCSI simply can't compete with WD740GD's simplified connectivity, relatively low price, and awesome (for a 10K drive) heat and noise profile, and SCSI's performance advantage over WD740GD is small to nonexistant unless you get an MAU or Atlas 15K II.

SAS brings simplified connectivity, but will probably set the price bar even higher for the foreseeable future, as the Adaptec 29160 and equivalent cards that frequent EBay won't be an option for cheap SAS entry like they were with paralell SCSI. SAS also won't magically make the drives run cooler/quieter.
 
doox00 said:
seagate charges you a 25 dollar fee to do advance rma
wow, that sucks
guess that's in the fine print when they're bragging about their 5 year warranty
 
well my replacement drive just failed :(
I'm beginning to wonder if I've got a bad power connector or port on my RAID card

what are the odds two different drives would fail in the same slot?
 
DougLite said:
I am extremely disappointed that WD is going to have to eat another warrnaty claim because of poor HDD shipping. Crap like that drives up prices for everyone :mad:

There ought to be a law! I really mean it.There should be some enforced standard for packaging.

Some people in Finland have told me that Maxtor drives are unreliable, but the ones that I've used seem to be bulletproof. My Maxtor 120GB 8M cache drive has survived 2 trips overseas in my suitcase, and diagnostics show it as being completely healthy. It makes me wonder if there are some drive handling problems over there.

Oh, by the way, the only etailer I've bought a hard drive from was bigfootcomputers.com (in Canada). When you open a package sent by them, you get swamped in a sea of styrofoam peanuts. They absolutely LOAD the box with those, you're faced with a big cleanup job after unpacking one of their shipments. ;)
 
Back
Top