Palimpsest found disk failure?

[pedora]

I have read some of the archived posts on this subject and was wondering if there are any updates.

I have the same message that my Drive is failing. I have a Seagate IDE laptop drive with 120GB capacity. I just upgraded my laptop to this drive from a 40GB HDD. I tried a Fedora LiveCD w/ Gnome. The KDE version didn't report anything so this utility is Gnome centric? Anyway, The stats display the status as 'failing' and 70 bad sectors.

Is this utility inaccurate, perhaps? There were some discussions and some people were suggesting not to worry too much about the results. But, this possible bug was present in Fedora 10? If it is a bug, then it appears it still exists in Fedora 11?!? Or do I have reason to be alarmed and fear possible problems with my hard drive?

Here is what ATA SMART Attributes shows (the 'failing' part)

ID Attribute ..... Status.........
5 Reallocated Sector Count 99 99 37 70 Sectors * FAILING

Comments?

[NoEffex]

Usually it shows errors that hurt during startup. Just my two cents.

[wangmaster]

All harddrives are designed with a certain amount of redundancy allowing the drive to reallocate bad sectors to good sectors. SMART has the ability to record when this happens.
Seagate and some manufacturers are claiming that as part of the normal manufacturing process, it's possible that drives might have bad sectors that are "pre-reallocated" for you and those will be reported i the SMART diagnostics which is what palimpset is complaining about.

There's a few schools of thought
a) a new drive should have 0 bad sectors, so if it's a new drive, you shoudl complain and try to get it replaced
b) some just accept the argument that some drives just come with bad sectors that are reallocated for you (you don't "lose" any disk space, it's just that there are some physical problems with the disk causing some of the redundancy to come into play)
c) if this is an older drive, and that number slowly starts to grow, then it means your disk is going bad

I find palimpset to be a little obnoxious in that it reports a disk as failing even if it's under the threshold. My external 1tb drive has 1 reallocated sector and has since i bought it a year ago, adn it's not growing so i haven't bothered to care but palimpset keeps on telling me it's a failing drive.

[jamesjpn]

My Seagate terabyte drive has only 31 days running time, but Palimpsest Desk Utility just reported a bad sector. I've never had a bad sector reported until Fedora 11 though I've been using Fedora since version three. This is the second time Palimpsest Desk Utility reported a bad sector. The first was an older 250 meg drive which was understandable.

I don't like to see that yellow triangle exclamation mark icon on my Gnome panel. Would it be OK to disable Palimpsest? I've never had a data loss yet due to drive failure and I have a good and regular backup system.

[glennzo]

To get rid of that annoying message (I suffered with it for a long time too) select System > Preferences > Startup Applications. Find Disk Notifications and disable that. You will not see the Palimpsest message any more.

[sideways]

I think palimpsest reporting sensitivity is under review. The sesnsible approach, as suggested above, would be an option to suppress messages until the "failed" count changes on a subsequent boot, rather than disabling disk notifications altogether.

I have an open bug report on nonsense (huge) values being reported, which is partly a bug in smartctl, this confuses the issue for me.

[JEO]

Someone opened a bug about palimpsest notification utility complaining about reallocated sectors and the maintainer set it to Closed: NotABug, said this behaviour was by design. Personally I don't want to be notified a drive is failing until the attribute value is at or below the listed threshold value (found by doing a smartctl -A command on the drive device). I would not be opposed to having it notify me that the reallocated sector count has increased since the last time it ran though. For now I just have the notification utility completely disabled.

[beaker_]

Having old junk myself, I find it anoying but useful. Ok I haven't seached for a man page, but I have googled the net for a manual to no avail. Isn't there anyway to configure this puppy? i.e., say this is I want palimpsest to do.

[Flanker711]

I bought a seagate 7200.12 500GB harddisk three weeks ago. Initially, it shows no error at all; however, it reported a bad sector since yesterday.

"Reallocated Sector Count 100 100 36 1sector failing"

I tried to do a selftest under Palimpsest. The "Extended" test never finished, I waited it for 3 hours but nothing happend, even the HDD LED does not blink regularly.

I have read some of the archived posts on this subject and was wondering if there are any updates.

I have the same message that my Drive is failing. I have a Seagate IDE laptop drive with 120GB capacity. I just upgraded my laptop to this drive from a 40GB HDD. I tried a Fedora LiveCD w/ Gnome. The KDE version didn't report anything so this utility is Gnome centric? Anyway, The stats display the status as 'failing' and 70 bad sectors.

Is this utility inaccurate, perhaps? There were some discussions and some people were suggesting not to worry too much about the results. But, this possible bug was present in Fedora 10? If it is a bug, then it appears it still exists in Fedora 11?!? Or do I have reason to be alarmed and fear possible problems with my hard drive?

Here is what ATA SMART Attributes shows (the 'failing' part)

ID Attribute ..... Status.........
5 Reallocated Sector Count 99 99 37 70 Sectors * FAILING

Comments?

[djs75]

As stated earlier in the thread - I'm not worried about the warning at this time as I have an older HD, but how can the threshold be set to a number that will be useful?

The man does address.

Thxs, for the disable method so I don't have to see it until it can be configured correctly.

[sadaltager]

Greetings.

I upgraded my HDD to a 250GB WD ATA only a month ago or so.
At the time I was using the latest update of KDE as a desktop environment and I didn't have Palimpsest installed on my system.
A week after the upgrade I switch completely to Gnome.
A few days after that the first disk utility warning appeared at start-up that showed only one sector failing.
Now there are four bad sectors, plus a considerably chunk of disk space (~3.4 GiB) that became unavailable.
Any thoughts?
Thanks.

P.S. Is it possible that the first bad sector occurred by forcing a downgrade of a package with the rpm command an the --oldpackage and --force flags.

[JEO]

If the reallocated sector count increases, then that means your drive is replacing the bad sectors with spares from a spare table. But when that happens you sometimes get a blank sector or a "hole" in your data which can cause random problems. For example, if the sector that was replaced was in a directory, you could lose access to many files or a file system area. If you lost a big chunk then you should run a file system check (fsck) on the affected volume, and the volume should not be mounted at the time so best to run it by booting from a rescue cd.

Downgrading a package would not cause any sectors to be reallocated. All of that is occurring inside the hard drive and its firmware is controlling that. General filesystem corruption could be caused by a bad package or a software error though.

[sadaltager]

Thanks a lot JEO !!!

[JEO]

You're welcome

[David Batson]

I have a WD160 5400 rpm PATA drive with 87.5 days on it according to SMART. Probably correct. Three or four days ago I get this 1 bad sector message with Palimpsest Disk Utility.

I have had Fedora 11 installed for about a month and a half. I also have Windows 7 RC1 installed in a dual boot configuration. Windows 7 RC1 is what I installed on the drive when I first got it. I ran a disk check in Windows 7, but it doesn't report anything. Of course Windows 7 doesn't support EXT4, so I doubt if a Windows 7 disk check could check the Fedora EXT4 partition for errors.

So I guess there is no way to mark off that bad sector in Fedora 11?
I believe Windows can do this.

Chkdsk (Chkdsk.exe) is a command-line tool that checks volumes for problems. The tool then tries to repair any that it finds. For example, Chkdsk can repair problems related to bad sectors, lost clusters, cross-linked files, and directory errors. To use Chkdsk, you must log on as an administrator or as a member of the Administrators group.

You can also run Chkdsk from My Computer or from Windows Explorer.

http://support.microsoft.com/kb/315265