Re: Corruption Stats: Suggested Blacklist from the data

Joseph Malicki (
Thu, 30 Jul 1998 01:08:29 -0400

Andrej Presern wrote:

> Does a bunch of "Deleted inode x has zero dtime" at fsck time count as fs
> corruption? I get these regularly when I force a fsck on an otherwise 'clean'
> fs. It so happens that I have a Quantum Fireball in the box and now that
> you've mentioned it I remembered the fsck complaints...
> And now that I started thinking about it, I also remembered that tar sometimes
> fails to unpack otherwise good gzipped tarballs (if I repeat the command,
> everything is (usually) swell). I was blaming tar/gzip for this but now I'm not
> so sure anymore (especially since I replaced both files and the problem still
> persists).

I'm not sure if this is related, but I've noticed that over non-PPP connections,
downloading files via ftp program often results in corruption, while ncftp
works fine, and I don't know why.... but I also get many "deleted inode
has 0 dtime" w/ DMA enabled OR disabled on IBM Deskstar 5 or
Maxtor DiamondMax 6.4GB (cant remember model #'s), which I'm
sure are fine drives.... could this be related to not flushing the IDE drive's
cache? I noticed NT 4.0 had a hotfix to enable flushing of write-cache,
so I am under the impression that it is possible.... and yes, my machine
does automatically turn off. But I have experienced no other fs
corruption except for locked kernels that I had to reset...

> 2.0.35 UP (Cyrix P166), IDE here..
> Andrej

Joseph M. Malicki

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at