Jump to content

lava890

Members
  • Posts

    11
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by lava890

  1. I’ve had this happen twice in the past couple years so I’d like to see what I can do differently:
     

    I have a pool on Windows 10 with about 10 drives. I have folder duplication enabled for several folders. I had a power failure, and it seems one of my 10tb drives got corrupted. When I started the server again, the drive was missing from the pool and was listed as an unpooled drive. There were no folders or files showing up in file explorer, however in drivepool it said that the 8tb that was previously being used by drivepool was “other” and not usable. In file explorer, the space was unavailable as well. I ran chckdsk /r for 15 hours and it found no problems. Scanner found no problems either. I think that file system corruption has occurred but there were no "found.###" folders.

    All of the files that were on that drive (duplicated or not) are missing from my pool. How can I prevent this from happening in the future? It seems like drivepool treats the duplicated files as though they were purposely deleted instead of making the pool “read only” until the missing drive is replaced or the duplicated files can be duplicated again. 
     

    As a side note, i have a cloud backup of the files and do not have to recover them. 

  2. 19 hours ago, Christopher (Drashna) said:

    Yes, please open a ticket at https://stablebit.com/Contact

    And ideally, attach drive tracing logs of this in action:
    https://wiki.covecube.com/StableBit_DrivePool_2.x_Log_Collection

     

    Also, it's very weird that Windows Defender would do this. as we do test with it. (well, it's hard not to).  But issues with BitDefender... don't really surprise me. 

    Also, in the meanwhile, it may be worth enabling the "bypass file system filters" option, as this should prevent the realtime scanner from being used when the pool reads the data from the underlying drives.

    It was sporadic, so it would have to be logging for several days to probably catch the problem, and it was pretty disruptive so I'd rather not turn the read striping back on since it's been working well now since July 30.

  3. 9 hours ago, DigitalPackrat said:

    Wish I had found this thread earlier, have been chasing a random corruption of movie files for over a year. It is so random and such a small percentage of files that I may not catch it for months till I am playing back a video file and find it corrupt at a random place, not the first 128MB though. Issue followed from Windows 2019 to 2022 server and a compete rebuild to newer hardware. Originally I was using SSD cache but moved to 384GB RAM with no change. Windows Defender is in use but drive pool is excluded from scans. Drive Pool is duplicating everything but read striping is disabled and verify after copy enabled now, hopefully that fixes it.

    Hope it works

  4. That’s possible, but I didn’t try read-striping with Defender turned off, so I can’t say it’s that combo and not some other aspect of my configuration. Either way, it caused me 2 years of frustration, but finally it’s solved!

  5. 8 hours ago, Jonibhoni said:

    Does the problem still appear if you disable Read Striping (Manage Pool -> Performance -> Read striping)?
    And do you have an anti virus software active that checks files on access?

    I have an issue with random file corruption when using the combination of both (Bitdefender + Read Striping), which is why I disabled read striping.

    When I turned off read striping, it solved the problem. I just use the windows antivirus which you can’t really shut off permanently. 

  6. 36 minutes ago, mhendu said:

    I ended up just moving to SnapRAID and this seems to have resolved the issue. The downside is there's no real-time parity, but I do like that I can configure it to scrub files routinely to detect corruption, which FlexRAID did not do.

    Ok, thanks for responding. I’ll experiment a bit more before I give up on drivepool. I think it worked flawlessly until I added the SSDs a couple years ago. 

  7. On 7/26/2021 at 12:22 AM, mhendu said:

    I have DrivePool installed on two computers - one where I've had it installed for quite a while with no issues, and another where I just finally migrated from FlexRaid pooling. I have the SSD Optimizer plugin installed but am using it to cache files on a regular 3 TB hard disk before moving them to my array (the other drives, but not this disk, are still set up with FlexRaid transparent RAID to provide some protection via parity).

    Although the speed of this pool is much quicker than with FlexRaid, I'm running into some disturbing file corruption issues. I've compressed a few movies with StaxRip and the resulting file, when placed on the pool, will have corrupt sections (note this is on the 3TB cache drive that is not set up with FlexRaid). Some of the movies worked fine when muxed to the cache drive, but then when file balancing runs they'll get corrupted when they get moved to a different drive in the array. This makes the product unusable.

    Any help would be appreciated.

     

    Did you ever figure this out? I’m having the same problem. Was it the SSD optimizer, read striping, the cables, bypass system filters, real time duplication, or something else? I’ve set pretty much everything to default, and I’m trying one by one to see what’s doing it, but it is sporadic so it’s tough. 

  8. 6 hours ago, Christopher (Drashna) said:

    The issues with USB are twofold: 

    • Shitty hardware used by various manufacturers
    • The USB spec allows for random disconnects, in the first place

    If you have an option, then eSATA is always the better option.  But if you have no option, then USB is okay. But I would personally disable write caching on the drives, or use ReFS to reduce the chances of issues.

    Thanks!

  9. I'm using a Mini PC/NUC (Beebox-S 7200U) as my Plex server with two external USB 3.0 hard drives in my Drivepool, and it has worked perfectly for the past year. I need more space, so now I'm looking at purchasing a 4 or 8 bay hard drive enclosure. I've read in this forum in posts from a few years ago that there are inherent issues with USB connections to these enclosures, and it's better to use eSATA. Does anyone know if the same USB issues are present with USB 3.1 Gen2 (Type C)? It's my only other connection type on this NUC.

    If so, and I have to re-purpose an old PC instead of using my NUC, is there a better connection type now instead of eSATA? I've read more recently that eSATA cards are having issues in newer Windows OS's.

    Thanks.

  10. On 10/1/2017 at 1:56 PM, Christopher (Drashna) said:

    Could you make sure that you're on the latest internal beta?

    http://dl.covecube.com/ScannerWindows/beta/download/StableBit.Scanner_2.5.2.3128_BETA.exe

     

    There are a couple of issues with the My Passport and similar drives that pop up in StableBit Scanner due to some issues with "drive geometry detection".  The linked version should correct that.

     

    If that doesn't fix the issue, let us know.

     

     

    As for the full format, it may be because it's allocating some of the volume information in a different way. Otherwise, not really sure (Alex is more of the expert at the low level drive stuff, than I) 

    I was having the same problem with a new 8TB WD My Book, found this post, and this seems to have fixed it.

×
×
  • Create New...