Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: mp3s getting corrupt randomly (Read 11963 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

mp3s getting corrupt randomly

Reply #25
While I still don't think the problems you are seeing with the disk are related to the corrupt MP3s (at least not all of them), bad sectors are never a good thing.  Backup your data as soon as you can.  If you've already got bad sectors there is a good chance you will get more as time goes on.  If the disk is new enough you should be able to RMA it.

mp3s getting corrupt randomly

Reply #26
Seek error rate seems not "OK" to me as reported by your tool. 3929993250278 (0x39305DA05E6) seek errors is quite much and looks rather broken.

EDIT: Just read that Seagate reports internal positioning errors (which is normal at todays densities) as SMART seek errors. So this metric may be no problem. But Reallocated sectors is a bad sign, as washu said. I don't think it's the mentioned trojan. Else you would get prompts to install a codec.

mp3s getting corrupt randomly

Reply #27
While it might not be the trojan mentioned, I'd still like to know if videos and other files as well are affected by this symptom. It would pretty easily rule out the possebility of eigher disk failure or software malfunctioning/virus etc.
Can't wait for a HD-AAC encoder :P

mp3s getting corrupt randomly

Reply #28
No, this is not a sign of a harddrive failing. Windows keep an MFT record of all existing files on the drive. This is obviously being tampered with or getting corrupt, but I disbelieve that it's caused by drive failure. Windows (and drives possebly too) checks for CRC during reading, and if it fails it will throw an error.


No, both of Windows' filesystems NTFS and FAT don't have CRC checking capability. They don't even record CRCs. In fact, most filesystems don't, only very advanced ones like Solaris' ZFS. Windows having a MFT doesn't have anything to do with it either. Approaching hard drive failure is the most probable cause.

To back up my theory, I accudentially stumbled upon this article from a recovering company in this case explaining a few things especially about Seagate (Failgate?) consumer drives.

Quote
In our lab we use special imaging hardware tools that are capable of reading raw sector data ignoring checksum check.


If anyone, these guys would know what they are talking about!
Can't wait for a HD-AAC encoder :P

mp3s getting corrupt randomly

Reply #29
Quote
In our lab we use special imaging hardware tools that are capable of reading raw sector data ignoring checksum check.


If anyone, these guys would know what they are talking about!


You are confusing filesystem checksums (which you claimed would exist in Windows) with a hard drive's internal, proprietary checksums (mentioned in the article). The former don't exist as often you may repeat the claim. The latter are not accessible by a host operating system.

A failed drive may still show unexpected and not defined behavior. For just one example take Slipstream's case. The only way to verify that your data is really consistent are filesystem level checksums (e.g. ZFS) or, if that isn't available, at least file level checksums (file formats with inbuilt checksums or separate parity files).

mp3s getting corrupt randomly

Reply #30
Hi I'm late to this thread (just had a brief read through) but have had a lot of experience with HDD problems. Here's my 2 cents worth.

1. I don't think this particular issue is caused by the HDD. Reallocated sectors are not the same thing as "bad sector" as reported by chkdsk for example. Since sectors are internally reallocated by the drives firmware the situation generally has to be pretty bad before it gets to the point where the OS actually reports bad sectors. In other words, bad sectors reported by the OS (eg chkdsk) are really a bad thing these days, but not really so much the drives internally reallocated sectors. Don't get me wrong, reallocated sectors are never a good thing, but with modern HDD’s having a few is not usually a sign of imminent disc failure, unless the number is increasing rapidly.

2. Even though I don't think this drive is currently failing I definitely think EvilOL is being way too cavalier about the need to back up. Trust me EveilOL, it doesn't matter what brand HDD you use, if you haven't had a catastrophic HDD failure yet then it's only because you haven't lived long enough. If you just use one or two PCs then yes you can be lucky and not see a single HDD failure for years or even decades, but if you look after a lot of PC's you soon come to realise that HDD's fail all the time, all brands- all the time ok. So backup or sooner or later you'll lose everything.

3. This type of random file corruption is far more often caused by ever a bad controller (ATA or SATA) or bad ram. Don't overlook bad ram, it's the number one cause or randomly corrupt files in my experience. Run a good memory test on the system. It also wouldn't hurt to update your chipset and SATA drivers to the latest, and also the system bios to the latest.

Of course I'm looking a hardware explainations. If you have viruses or malware then that's a whole other thing.

mp3s getting corrupt randomly

Reply #31
2 cents about RAM:

1. Defective RAM usually correlates with a generally instable system.

2. Data files can only be corrupted at writes, e. g. when a file is initially written to disk or when defragmenting. Just reading files shouldn't corrupt their contents. The MFT could become corrupted, though, but that shouldn't modify file contents, just make them inaccessible without further measures.

Controller/mainboard/driver issues also seem most probable to me, now that the reallocated sector count is so low (although this is a sign that the drive may die soon due to microscopic, scraped-off particles from the platter).

3 cents about malware:

What people generally don't get about viri and trojans, is that they usually all serve a purpose. Either they intend to control your system and open a channel for the reception of further code or they encrypt your files to extort ransom. The only known case that modified mp3s on your disk tried to make the media player download additional "codecs" by modifying the internal filetype. Programs that solely damage or delete your beloved files are mostly fiction. As in nature you can't use the host for further reproduction if you kill it. Also when modifying files you want to make sure that the user isn't alarmed because they don't work anymore.

mp3s getting corrupt randomly

Reply #32
Controller/mainboard/driver issues also seem most probable to me, now that the reallocated sector count is so low (although this is a sign that the drive may die soon due to microscopic, scraped-off particles from the platter).

Yes that's what I meant about watching out for a growing reallocated sector count . When I notice some reallocated sectors I like to do a "chkdsk /r" (surface scan) and see if either the reallocated or pending count rises. If it doesn't rise then the situation might be stable and the drive might survive a good while, but once I've got reallocated sectors I definitely like to keep an eye on them.

mp3s getting corrupt randomly

Reply #33
Quote
In our lab we use special imaging hardware tools that are capable of reading raw sector data ignoring checksum check.


If anyone, these guys would know what they are talking about!

You are confusing filesystem checksums (which you claimed would exist in Windows) with a hard drive's internal, proprietary checksums (mentioned in the article). The former don't exist as often you may repeat the claim. The latter are not accessible by a host operating system.

I did mention "and drives possebly too", didn't I? So maybe NTFS doesn't have any error-detection itself, but I take this would usually be sufficient. Don't tell me that I'm confusing something when I explicitly point out what I mean. The only thing I've confused here is that NTFS might not have error-detection built it. Reason I believe it did, is that I have seen Windows throw errors when it cannot copy data, and I have never seen data become corrupted all by itself or by copying. However the error I'm reffering to, probably came from a CDFS filesystem. I suppose it has CRC checking due to the higher risk of media failure.
Can't wait for a HD-AAC encoder :P