Everything first fell apart one morning during my automated backups. This is one of five disks in my Ubuntu 11.04 linux box. It's part of a 2-spindle LVM stripe which is normally offline, but which gets mounted right before the backup and then unmounted immediately afterwards. Mid-way through the backup, the filesystem on that backup partition disappeared, and I was unable to remount it. My kernel log (/var/log/messages) contained the following errors:
kernel: sd 12:0:0:0: [sdi] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
kernel: sd 12:0:0:0: [sdi] Sense Key : Aborted Command [current] [descriptor]
kernel: Descriptor sense data with sense descriptors (in hex):
kernel: 72 0b 47 00 00 00 00 0c 00 0a 80 00 00 00 00 00
kernel: 36 71 d9 c0
kernel: sd 12:0:0:0: [sdi] Add. Sense: Scsi parity error
kernel: sd 12:0:0:0: [sdi] CDB: Read(10): 28 00 36 b1 d5 00 00 00 80 00
and a few minutes later:
kernel: sd 12:0:0:0: [sdi] Unhandled error code
kernel: sd 12:0:0:0: [sdi] Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK
kernel: sd 12:0:0:0: [sdi] CDB: Write(10): 2a 00 05 54 c0 00 00 04 00 00
kernel: lost page write due to I/O error on sdi3
and finally, the kicker:
kernel: sdi: detected capacity change from 1500301910016 to 0
Yup, the drive capacity changed from 1.5TB down to zero bytes. Time to replace the drive. Fortunately, it was still seven months shy of the end of its three year factory warranty.
Samsung's web site informed me that their hard drive division had recently (as in, they haven't quite smoothed out all the kinks yet) been purchased by Seagate Technology, and pointed me to their web site. Seagate's web site instructed me to run their SeaTools diagnostic software, which would give me a test code indicating the type of failure. I let it run for 9 hours over night, and the progress bar still showed 0%. During that time, it found over a hundred bad sectors on the drive. I killed the tool (a boot CD), and rebooted the rest of the computer without that drive.
I decided to call Seagate to see if (A) I could have them ship me the new drive first so I could get back in business sooner, and (B) if I could pay the difference to upgrade to a larger & faster drive instead of getting an identical replacement. Seagate's customer support web site is fine if you want to download product information, but heaven help you if you want to interact with a human. Even their "contact us" web page lists only a snail mail address. No email, no phone, no live chat. Just snail mail.
By googling for "Seagate customer support phone number," I found that I wasn't the only person who was incredibly frustrated at this -- and it had been that way for at least 18 months. I was, however, able to find through a 3rd party forum the URL for an online chat line for Seagate tech support. Of course, that chat line doesn't support Firefox, the most common browser in use today. I dug up a Windows computer with IE8, only to find that their entire support department shuts down for a 90-minute lunch between 11:30-1:00 Central Time. It was 11:42 AM.
Later that afternoon, I was able to chat with Thomas, who verified that my SeaTools behavior was indeed indicative of a dead drive, and that I wouldn't really need a test code in order to process my warranty. He couldn't help me with my other warranty questions, because that wasn't his department. He was able to give me a phone number for their customer support: 800-SEAGATE (800-732-4283).
On the phone, Clarence said that because this is a Samsung drive and they haven't quite finished integrating the Samsung devices into their Seagate system, I would not be offered the opportunity to upgrade my new drive to something a little faster & larger than the 2.5-yr-old drive that just died. He also said that, for the same reason, they couldn't offer advanced replacement, which means that the won't send my new drive until they receive my old one. They estimate 7-10 business days for the whole transaction. Oh, and they also wouldn't pay return shipping on my dead drive, even though it was under warranty. This means I have to wait 2 weeks and still pay money to get a working version of a slow, small hard drive. After just a little complaining, Clarence did agree to email me a pre-paid UPS Ground shipping label, so at least the new drive will be free (though still delayed).
I boxed the drive up that night & sent it on its way the next day. According to UPS, it took the drive 5 calendar days (including a weekend) to reach Seagate. It took Seagate another 4 days to log the drive into their system and send me a "we got it" email. It took yet another 6 calendar days (4 work days) for them to ship out the replacement drive. All told, I was without my primary backup drive for about three weeks despite the warranty.
I'm really bummed that I wasn't able to upgrade to a larger drive. That backup partition was over 80% full, so I'll need more space before too long. I don't have any more empty SATA ports on my mobo, otherwise I'd just buy yet another drive to add to the stripe for increased speed as well as space. I'll cross that bridge when I come to it, I suppose.
For the record, Western Digital didn't allow me to upgrade drives during that warranty repair, either, but at least they were easy to get a hold of and didn't waste much of my time. If this experience is typical of Seagate customer service, then I don't think I'll ever be buying another of their products.
As always, if you've got any comments or questions about this mess, please speak up in the comments section below.
No comments:
Post a Comment
Please leave your comment below. Comments are moderated, so don't be alarmed if your note doesn't appear immediately. Also, please don't use my blog to advertise your own web site unless it's related to the discussion at hand.