9.10 upgrade says I have failing hard drive

[ScottinSoCal]

I'm thinking bug.

Normalized: 97
Worst: 97
Threshold: 24
Value: 6029404 sectors

Running on a Dell Latitude D810, ATA Fujitsu MHT2080AH HDD

I still ordered a new HD for it. Upgrade from 80GB to 160GB for $100. Not bad.

[Devi 710]

Doing a long test (including repairing) with "Seagate Seatools for DOS" solved my problem!
Actually... it reduced the number of sectors "waiting to be reallocated", so, no more warnings for my Seagate drive

Any idea how to get the Seagate Seatools for DOS booting from a USB? I am using a netbook so I don't have a CD Drive and it won't boot from a USB.

My threshold is "0" and the value is "153" (up from 60 the other day)

[deamon_knight]

This should be considered a bug. Palimpset may be accurately reporting how many reallocated sectors you drive is reporting, but forming the wrong conclusion from that data You only have a failure when you run out of reallocateable sectors. Even with some sectors that are bad and cannot be reallocated a drive may functional without observable difficulty or data loss. What to take away from this is that all rives will fail and with little real warning, so you should make sure you have any crucial data backed on some other disk/media so you can survive a failure.

In my case, my Western Digital HDD passes all of the Manufacturers test and reports Normalized SMART Values, of 200 with a threshold of 140, Values Under 140 are reported as errors. (It does not report the RAW data as Palimpset does)

Palimpset warns that I have 183 pending (and that there are 1982 bad sectors, an amazing number). This is not reallocated or bad sectors but sectors that need to be checked to be reallocated next time data is written to them. If they write correctly they are flagged good and are reallocated if they fail to write.

Ultimately this means nothing more than I might be having a failure soon, maybe, so its not diagnostically meaningful. The bug is that Palimpset set is reporting sectors flagged as "Possibly Bad" to be "Definitely Bad", when this is not the case.

To satisfy my own curiosity I'm going to zero out this drive and see what happens.

[deamon_knight]

Yea, I completed a full wipe with Western Digital's software utility, wrote zeros to entire disk. After I reinstalled 9.10. Palimpset now no longer reports imminent disk failure.

The Category I had was "Current Pending Sector Count", and it no longer shows 1900+ bad sectors. I have no idea what this value is supposed to record. I gather it means the HDD firmware has identified some sectors as "for review" and wants to check for correct behavior next time data is written to those sectors. Since most users aren't writing over large sections of their disks on a regular basis, these sectors never get written too and thus were never written/checked until I wrote zeros to the entire drive. Palimpset's bug is reporting these as "bad" (failed sectors) , when they are not.

[phillw]

palimpset is the subject of a bug report for false positives https://bugs.launchpad.net/ubuntu/+s...rt/+bug/438136

For those affected, do take the time to get a login name & report what it is reporting - it is an 'active' bug and you can follow its progress there.

Regards,

Phill.

[Sentience]

Has any9one gotten their hard drive fixed after having this problem?

[Cr125f150ba]

I'm also thinking this is a bug because I ran a memory test on my hard drive and it said it was fine. Then just to make sure i installed Windows XP and checked the status of my hard drive and it also said my hard drive is fine. But when i installed Ubuntu 9.10 It always says I have many bad sectors. So I'm pretty sure it's a bug because I'm sure windows xp would have caught it and the memory check i ran would have found some errors if my hard drive really had bad sectors.

[ScottinSoCal]

It was not a false alarm, although I still don't believe the numbers it was reporting. TigerDirect sent me a 320GB hard drive, for the same price as the 160GB - I guess they were out of the one I ordered. Yesterday my computer started locking up randomly, and the HDD was on the way out. I replaced it, reinstalled Ubuntu, and managed to get my key files off the old hard drive. It's gone to Hardware Heaven now.

FEBE got my bookamrks, user IDs and passwords off my WinXP installation and transferred to the new Ubuntu installation. While I was at it, I researched WINE and successfully transferred my Forte Agent to Ubuntu, and this allows me to put my WinXP laptop into storage. The only things I still need it for is Spore and Bryce. So, a failed hard drive turned out to be a good thing.

[Jakxx]

I had the same problem when I upgraded. I bought a replacement hard drive and received the exact same error again, so I can only assume that it is a bug.

[BeanBagKing]

Just found this thread, though I'd leave a comment on my experiences.

Was using XP on this laptop, which was running horribly slowly. So I installed Ubuntu and got the "DISK HAS MANY BAD SECTORS" warning. Well, I figured I found the problem and used Ultimate Boot CD to run SeaTools (Seagate Hard Drive) and run an extended test. Sure enough, got a bunch of failures and write errors.

Then I came across this thread looking for more info where a lot of people are saying that the "Disk has bad sectors" message in 9.10 is just a bug.

Moral of the story here is that for some this may be a bug with 9.10, but in my case it was correct. Don't read this thread and simply shrug it off, run some other tools (especially the manufacturers) and see what you get. Even if all this is good, backups are always smart.

For now my disk is still working fine, I never would have known had it not been for 9.10 until a catastrophic failure and everything was lost. Just because your disk "acts" fine doesn't mean it is.

If your getting this warning don't ignore it! Even if it turns out to be nothing, it's hard to be sure with a failing drive.