Writing speed slows down and freeezes (perhaps with large volumes?)
Open source disk encryption with strong security for the Paranoid
Brought to you by:
idrassi
Menu
▾
▴
In the past, I have used approx. 10TB and 18TB Veracrypt volumes, a couple of years ago when I changed NAS, I created 40TB NTFS volume (yep, mainly using Win10).
Slowness when writing to the volume:
For some reason, recently the volume in question has seemed slowing down as it has filled up when I've transferred larger amounts of files, i.e. ~500GB at a time. Unfortunately, I haven't measured this.
Freeze when writing to the volume:
Now I noticed while doing the transfer that the data transfer freezes when writing more data to the volume (tried several times and the freeze occurs when I have transferred approx. 2-3GB of files there). This does not happen when testing the transfer of a few smaller files (200MB x3). The volume still has more than 4TB of free space, so that's not the issue. There is no problem when transferring ~500GB packages for other parts of NAS than that Veracrypt volume.
Reading works fine from the veracrypt volume:
That is, when I try to copy files off the Veracrypt drive, there is no problem here, and I have not noticed any problems with the reading speed.
Has anyone else had similar problems? And can the problem be in NTFS etc. and not in Veracrypt?
Last edit: Aapole 2023-06-14
In my experience, a drive that starts exhibiting slow writes is an indicator the drive is starting to fail.
Check the Windows Event Viewer around the timestamp of when the freeze occurs. You may see errors being logged for the drive.
Thanks Enigma2Illusion,
Event Viewer shows that:
Event ID 137 NTFS The default transaction resource manager on volume X: encountered a non-retryable error and could not start.
Event ID 50 NTFS {Delayed Write Failed} Windows was unable to save all the data for the file \x.x. The data has been lost. This error may be caused by a failure of your computer hardware or network connection. Please try to save this file elsewhere.
NW is already tested (different route), next step is to test all drives within raid pack. Thanks for pushing me to the right direction!
The errors you posted mean your drive has failed and data corruption is likely for this drive.
Depending on your RAID enclosure, you need to eject this drive from the RAID and replace the drive.
Thanks Enigma2Illusion,
No errors were found on any of the disks after running the long tests on the disks - and detailed review of the smart data of the disks after that, still nothing indicating errors. I run also data scrubbing on the drive (btrfs) where the aforementioned volume is, just to see if it founds any mistakes. Also went ok. These took a few days to drive. So at least it seems that the disks are ok.
It seems that it is not a problem with the disks, but with something else, possibly with NTFS?
When running Veracypt's Repair Filesystem; Stage 2: Examining file name linkage; after it runs for a while, a message appears twice: an Unspecified error occurred... and the operation stops there.
Tests of larger amount of data against backups (few TB's), all ok. Deleted few TB's off the volume, Tried after this if I could move data there again. The same situation, approx. 3GB's of data is transferred, after which it freezes. The freeze goes away when I unmount the disk. And then again, it's possible to transfer ab. 3GB's of the data. Weird. Tested with another computer, same happens.
I'll keep trying in the next few weeks to see if I can find a solution to this, because it would be always nice to know what's the cause of an error > if I find it, I'll of course report it here. If no solution is found, I delete the volume and create a new one + restore from backups.
Related to thread here:
https://sourceforge.net/p/veracrypt/discussion/technical/thread/76027b406c/
and only NTFS it seems.