lkml.org 
[lkml]   [1999]   [Aug]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] roubust ext2fs against failure
From
Date
Hello,

> Anyway this is the kind of patch that I'd really like to see
> comments on from Linux big names. Either it's conceptualy

I'd like to too.

> Lastly, I bet you'd better refine the sentence:
> 'Just mount ext2fs with "sync" option, ext2fs will keep
> consistency of filesystem. Your filesystem would never be broken.'
> in http://www03.u-page.so-net.ne.jp/da2/h-takaha/userguide.html
> Do you mean that you force the controler and disk to write meta
> datas in the right order to the disk, or do you assume that
> there is no order problem at controler and disk cache level ?

Yes you must be right, every disk has a cache in itself.

It dosen't cause any problems when it happens to crash _not_ depeding on
power-failure, all metadata on disk-cache (which is in the disk) will be
write to the disk.

A problem is a crash depeding on power-failure.
Some disks may keep the order, some may try to flush all disk-cache,
others may not do that might cause little corruption.

I think journalling filesystem like XFS has the same problem.
Writting order of "journal" and "metadata" on disk-cache to real-disk
might be changed in some situation, it would cause a filesystem corruption
which can't be detected at journalled recovery after a crash.

It seems to be not good idea to configure disk-cache as write-through-cache
that leads a poor performance.
Is there any cool solution?

Regards,
Hirokazu Takahashi

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:53    [W:0.045 / U:0.020 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site