My dear and respected HDD gurus, I am the owner of Seagate Barracuda ST3500418AS 500GB, from 7200.12 series, fw CC34, which failed several days ago. BIOS doesn't detect it, yet it tries really long as if it knows that some drive actually IS connected to SATA controller but it doesn't respond well, so it gives up after a while reporting no drive at that particular port. My suspicion is confirmed by kernel (yeah, I'm a linux user) which bypass BIOS and also detects that some ghost drive is connected to controller and tries to communicate with it but gives up after several retries claiming “...the link is too slow”. Freeware diagnostic tool MHDD also seems to correctly suppose the drive is connected to SATA port, but after several retries it claims 'Drive not ready'. Seatools doesn't detect it at all, probably because it relies on what BIOS says.
It is now around 6 months that I got this drive as a replacement for a well known failure-prone 7200.11 drive. I waited almost a year before I exploited warranty since data stored on that 7200.11 was sensitive so I let time to pass by until data became valueless. Fearing that something nasty would also happen to new drive (after all, the difference from the previous is just 0.01), I didn't put any sensitive data on it. Moreover, after three months I've noticed unusual (read: stupid) things happening to the replacement drive. It would be spontaneously reset during work to lower speed, with kernel claiming that ATA link became slower. Immediately before that system would usually freeze for around 10-15 seconds. Sometimes, after resetting the drive, kernel would mount file system as read-only claiming it needs checking. After checking, there would be a pile of bad sectors, usually around 30-40. Sometimes, after booting again, the drive would pass undetected by BIOS, but hey, it would resurrect itself after 2nd or 3rd or even some higher order booting. Sometimes, BIOS would report it has 0 GB capacity after rebooting. And guess what, all those cordons of bad sectors would disappear after several bootings and the file system would now be clean as popes dress (that's why I didn't send it to be repaired, fearing that it would present itself as a perfectly healthy disk to a servicegirl and make me an idiot – moreover SMART tests and SEATOOLS always reported the drive was doing great). When it happened last time, some month and a half ago, I wiggled the SATA cable and noticed that it was not-so-tightly connected so I thought to myself: HOORAY!! That little untihgtliness surely MUST be the reason for the drives malfunction. And really, after I pushed my connector more deeply to her socket it worked flawlessly until several days ago. When it happened again, I checked the cable, but it was tightly connected this time. So, bearing in mind past reincarnations, I was sure it would as usual arise from the death. However, it didn't.
I just wanted to ask You, my respected and well informed HDD gurus, does this strange set of behavior looks familiar to You? Do I have a justifiable reason to nourish my hopes that the drive can be repaired or should I bury them together with my memories of SEAGATE? The only reason why I would want this particular drive back from repair is because of not so relevant data which is stored on it. If it is some well known issue which can be fixed easily, then I would like THIS drive back, but I definitely don't want any new replacement drive (except if SEAGATE ships me WD or HITACHI, I mean there must be some disillusioned me-like WD user who is passionately dreaming of SEAGATE, so WD and SEAGATE should make a deal about switching the replacements in these cases). But in any case money return would be mostly welcomed (even without naturally expected compensation for loosing my nerves).
Thanks for any (even unusable) advice.
