Jump to content
  • 0

File evacuation, does this plugin move corrupted files?


manticore.mm

Question

At the moment stablebit scanner detected an unreadable sector on one of my 1,5TB Samsung drive.

 

I dont know witch files are damaged because scanning disk surface is at 97,41% and it says "you will be able to start a file scan when the disk surface scan completes" -> this is ok.

 

Surface scan has stopped because the plugin is evacuating the files from this disk (99% duplicated files)

 

So what does happen with the corrupted files? Are they evacuated in that corrupted state to an other disk? This would be bad, because of duplicating i have an healthy copy of these files. I will never find out what was the healthy or the corrupted file?

 

If this is true, it is a really bad default setting for stablebit scanner plugin to move duplicated files from damaged disks

Link to comment
Share on other sites

5 answers to this question

Recommended Posts

  • 0

Hi

 

"So what does happen with the corrupted files? Are they evacuated in that corrupted state to an other disk? This would be bad, because of duplicating i have an healthy copy of these files. I will never find out what was the healthy or the corrupted file?

 

If this is true, it is a really bad default setting for stablebit scanner plugin to move duplicated files from damaged disks"

 

In my experience of having a corrupt file either by bad sector or bitrot duplication is useless. which has annoyed me to no end even when scanner tries to repair the file it doesn't, I have brought this up lots of times asking to have the option to replace the damaged file with the duplicate or for drivepool/scanner to have a similar setup like storage spaces where files are automatically repaired using the duplicate.

 

I don't think it's at the top of the list of things to do however it should some have some kind of priority since this is user data at risk but Alex is neck deep with cloud drive and everything else is on the back burner so I wouldn't hold your breath.

 

I have been tempted to not use dulplication and rely on backups which would save me about 12tb of space but because of how drivepool spreads the files across multiple drives and I have been unable to find any software that will scan the individual drives and create a list of contents that is easy to read it's easier to keep duplication on.

 

I do run a separate pool using storage spaces (refs)primarily for testing with 4tb of data because the self heal abilities appeal to me and as yet I have not had any loss of data as far as I know if Alex could add such a feature to drivepool i think it would add a lot more appeal to drivepool

Link to comment
Share on other sites

  • 0

hm i checked the manual section of drive pool and it says:

 

Protected files are never copied from the drive that's being removed. Instead, they are regenerated from one of the drives that holds the other duplicated file parts.

This is important when you're removing a drive that's already damaged, as the integrity of the drive will not affect duplicated files.

 

This is ok. So i think it would be the best to setup the scanner plugin to only move non protected files away from the damaged disk.

And then, if you remove the damaged disk, on this process the health copy will be duplicated again.

 

I'm right?

Link to comment
Share on other sites

  • 0

You can stop the surface scan, but clicking the "Stop check" button in the toolbar at any time. However, note that this disabled the automatic scanning, and you'll want to enable that latter. 

http://stablebit.com/Support/Scanner/2.X/Manual?Section=Toolbar

 

Once it's stopped, you should be able to run the file recovery scan.  This specifically identifies what files appear to be using the affected sectors on the disk. 

http://stablebit.com/Support/Scanner/2.X/Manual?Section=File%20Recovery

 

 

As for the moving the files out, this is normal and expected, as it is an attempt to prevent issues with any more files, as the damage sectors indicates a serious problem. 

 

 

 

So what does happen with the corrupted files? Are they evacuated in that corrupted state to an other disk? This would be bad, because of duplicating i have an healthy copy of these files. I will never find out what was the healthy or the corrupted file?

 

If this is true, it is a really bad default setting for stablebit scanner plugin to move duplicated files from damaged disks

 

These files are not necessarily corrupt.  Specifically, the damaged sectors are sectors that the disk is having problems reading from.  

Corruption indicates that all the parts of the file is readable, but has been altered.  

 

These are two entirely different things. And should be handled in different ways.  

 

That said, any files that use data in the damaged sectors will likely stall out when trying to read from them.  In fact, you *may* have seen this (either on the pool or not).  Errors like "Semaphore timeout" or other issues reading the files. 

 

 

Also, for StableBit DrivePool, when the file is accessed, we do check the modified date stamp. If that doesn't match, we verify the CRC values of both files. If that doesn't match, we flag the file as "file contents mismatch" in the UI and prompt for resolution. 

 

Additionally, when removing a drive from the pool, it will error out and stop when it can't read a file.  The "Force damaged drive removal" option will skip these files and continue to remove the drive.... leaving the damaged files behind. 

There is also a "duplicate data later" option, so you can quickly remove the drive, if a lot (or all) of the data is duplicated. 

Link to comment
Share on other sites

  • 0

thanks for your answer.

 

so whats your "best practice" for situations like this?

 

leave the option "move duplicated files from damaged disks" on?

 

i let the balancer do the job and the drive is now empty without any problems, but the file scan is not possible any more.

 

i am trying to find out whats the best way to deal with such errors like current pending sectors. if one drive dies completely, ok you have to remove it and replace it. thats easy. but what to do with drives showing few unstable sectors? when can you trust them again? never?

 

i removed the drive from the pool an now i write zeros to it.

Link to comment
Share on other sites

  • 0

Leave the option "move duplicated files from damaged disks" on?

 

i let the balancer do the job and the drive is now empty without any problems, but the file scan is not possible any more.

 

IMO (biased, of course) is to let it complete and evacuate the disk.  

 

Generally, once you start getting SMART errors, the drive isn't going to get better. 

 

 

i am trying to find out whats the best way to deal with such errors like current pending sectors. if one drive dies completely, ok you have to remove it and replace it. thats easy. but what to do with drives showing few unstable sectors? when can you trust them again? never?

 

i removed the drive from the pool an now i write zeros to it.

 

 

Current Pending Sectors usually can be fixed by writing to the drive. The drive will either continue using the disk, or remap/reallocate it.  The same *should* be true for uncorrectable sectors. 

 

However, IMO, once you start seeing these, it's time to get rid of the drive. 

 

I maybe a bit overzealous about this, as these errors are exactly how the Seagate ST3000DM001 drives start to fail.  And in fact, in a few cases, the drives failed minutes after finishing removing the content from the disk.  

 

 

 

 

so whats your "best practice" for situations like this?

 

Obviously, my answer is going to be biased here...  

Though, many can attest that I recommended and loved Scanner before being hired on...  

 

My recommendation is ... a bit complicated. 

 

Run Scanner for a bit, and make sure that you're not getting overheating warnings, as these are considered SMART Warnings as well.   Use the Heat settings to tweak the threshold, and the disk settings to override, if needed.  

 

Once that's done, enable evacuation on SMART Warnings for the pool's balancer.  This way, heat issues don't trigger this behavior (unless they're extreme).   Enable this for at least unduplicated data, so that you reduce the risk of data loss.  Optionally on duplicated data, as well. 

 

 

RMA any drive that has any SMART warnings (except temperature).   

Buying drives that have longer warranty periods may be worth the premium, as it adds additional piece of mind when dealing with drives. 

 

For unreadable sectors (damage), confirm that it's not a communication issue (mark as unchecked and scan again, and run the burst test).  If it's damaged, RMA (use "bad sectors" as the cause) the drive.  

 

RMAing is there for a reason. Take advantage of it.  

In fact, if you check Disk Details in StableBit Scanner, it may list the typical warranty period for that model.  This is from the data of manufacture in some cases, and not the purchase date, so be careful.  Also, it may be worth checking specifically for each drive, so you know when their warranties expire. 

 

 

Also, if you have the funds, ALWAYS do the advanced replacement. WD/Seagate/etc puts a hold on your card, and sends you the drive first.  You usually have a month to send back the problem drive, or they charge you the MSRP of the drive.  

 

Since it makes replacing problem drives faster and easier (as you can have both the original and it's replacement connected at the same time), it's worth the hassle, IMO. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...