Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #82281 > unrolled thread

Bug#998627: linux: please enable the new NTFS3 driver in 5.15

Back to article view | Back to linux.debian.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Bug#998627: linux: please enable the new NTFS3 driver in 5.15 Viðarr <vidarr@mailbox.org> - 2024-04-20 23:10 +0200

#82281 — Bug#998627: linux: please enable the new NTFS3 driver in 5.15

Hey folks,

just to chime in. I had been using the commercial driver from Paragon 
without issues _before_. I switched some of my machines to Manjaro and 
and kernel 6.6. So I "had" to switch to ntfs3, given their commercial 
driver won't build against the 6.6 kernel.

Boy, was I surprised to find that the ntfs3 driver corrupted a partition 
(see forum thread here: 
https://forum.manjaro.org/t/re-ntfs3-keeps-corrupting-my-ntfs-partitons/157736). 
Nothing that their _commercial_ chkntfs wasn't able to repair (the 
userland tools work, but their commercial driver lags behind in terms of 
linking to the latest kernels), but that's: DATA CORRUPTION.

Performance in terms of speed is nice to have, but data integrity is a 
more important performance feature for a file system; at least in my book.

So, my confidence in the ntfs3 driver has been shattered for now, I have 
to admit. I had in the past used the commercial Paragon driver on some 
of my machines, but ntfs-3g where ever I didn't have a license. So I 
have worked with both before. The better tool support from Paragon and 
the performance improvement were the main selling points to me, compared 
to ntfs-3g. I never encountered data corruption with the commercial 
driver or ntfs-3g.

But the ntfs3 driver evidently has such issues. And while those quoted 
threads are old, I think there is more recent evidence that there are 
issues ...

Citing the stability of their commercial driver makes no sense, by the 
way. They don't share the same lineage. Their Linux driver is - 
according to them - a rewrite, whereas the commercial one traces its 
roots to NTFS for DOS.

In light of my experiences with ntfs3, I can understand Salvatore's 
skepticism towards that driver.

Just my two cents,

Viðarr

PS: the only thing that can be said in favor of enabling the driver 
would be that it enables a larger user base to test it and weed out 
issues. But I wouldn't call its current state stable or robust, given my 
experiences and those that turned up in web searches once I started to 
look for solutions.

[toc] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web