Tick tick tick... significant number of Seagate hard drives failing | Hacker News Reader

Tick tick tick... significant number of Seagate hard drives failing

tuaw.com

14 pointstransburgh14 comments

Hm, I was originally buying exclusively Seagate due to their excellent warranty policy. Well, I thought it was excellent until I had to RMA a drive. They expected me to ship it to them in my own packaging where as WD sent me a box+foam first for free, with no questions asked. Seagate wanted $20 for their "+1" package that sent you a new drive first and asked you to send you the old one in the packaging they used.

I bought quite a few Seagate drives back when they were the first to move up to a five-year warranty; I believe it was around the 7200.7 series. Those were great drives - fast and reliable (knock on wood, I don't think a one has died on me yet).

Shame to hear about the .11s. It seems odd how cyclical the hard drive industry is... company gets popular, has a massive defect, consumers run to alternatives, until they have a massive defect and everyone flocks back.

My dad had a comment about this. When buying hard drives for work (a university), they would switch to whichever one didn't have a large failure.

They exhausted the space of the industry reasonably quickly. Just goes to show that for the (reputable) companies, major failures are pretty much random (although it might be nice to see some statistics?) and thus it's a wash.

I bought 200GB Seagate Baraccuda a couple years ago and have since gone through 2 RMAs. Though the hard drives fail, the RMA process is one of the most fluid and painless RMAs I have encountered.

The tools to check your drive are windows only.

Use hdparm -i /dev/hd* for ide internal

check /proc/scsi/usb-storage for external to see if you're affected in linux.

The article explains how to check in OS X using "About this mac". Sadly, you will need windows to actually update the firmware.

Good thing I have a backup drive ... which is ALSO Seagate.

Before panicking, it is worth noting that Seagate says [1] "Seagate has isolated this issue to a firmware bug affecting drives from these families manufactured in December 2008"

[1] http://seagate.custkb.com/seagate/crm/selfservice/search.jsp...

On a tangent, does anyone know how to perform recovery on a HDD that is not detected by the bios (e.g. due to defect/damage on the pcb)?

I have a 500GB IDE Hitachi Deskstar drive (was originally in an external USB caddy, but now taken out) that I think has suffered overvoltage. Even though there are no visible burn marks on the PCB, when connected directly via IDE to any PC I try, the PC will not powerup let alone POST, which indicate some short-circuit perhaps? This happens even if IDE cable not connected, which leads me to think of overvoltage damage (could be wrong).

Replacing the PCB with one from an identical drive does not work due to the fact on most modern drives the PCB microcode is unique to the specific unit itself (swapping PCB with an identical unit [make, model, batch, MLC, all same] results in drive spinning but no detection in BIOS or even by the manufacturers own drive fitness tools).

There are companies that will get the data off the drive for you. The prices for this kind of service have come down quite a bit in recent years. Backups are cheaper though...

I know there are drive recovery services, and have backup of most of the data, it is the challenge of fixing it.

Anyone remember the IBM "Deathstar" drive issues? Think that all but killed their consumer drive sales.

I read about this on heise the day after I placed an order on a 1.5TB Barracuda 7200.11. Hope they manage to get an efficient firmware update process in place - having to email support individually with the serial number is just not going to work.

http://jungledisk.com (not affiliated at all)

I was going to buy one. Quite glad I chose Samsung instead.

Reply on news.ycombinator.com