[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: XFS in Fedora Core 2



On Sat, Dec 20, 2003 at 12:11:03AM +0100, Xose Vazquez Perez wrote:
> Alan Cox wrote:
> 
> > Hans Reiser doesn't like Red Hat. Quite why he thinks a vendor should take
> > arbitary patches for an fs that isnt heavily supported and may break fs's that
> > are is a bit of mystery
> 
> nor ext3:
> 
> "ReiserFS V3 in the official kernel is much more stable than ext3 for a simple
> reason: we shipped first, and had longer to stabilize. I would say that
> historically we have always been more stable than ext3 was at the same moment
> in time in the official kernel, but that both filesystems were unstable in their
> beginnings."

Since Stephen is too gallant to take the bait, let me say a few words
without rekindling flamewars long dead.

1. Since it's very first alpha release, Ext3 treated user data as the
   precious commodity that it is.  Don't take my word, it is plain
   to see in the ext3-users archive, or the assertions in the code.
   Andrew Morton and Andreas Dilger continued that tradition with the
   initial 2.4 port.  It took a long time to get ext3-2.4 going because
   (1) Stephen was busy fixing the core kernel, and (2) recognizing
   that Ext3 would be the de facto replacement for Ext2, extreme caution
   was exercised.

2. In the beginning, Reiserfs lacked a fsck.  For the longest time, it lacked
   a useful fsck.  The method of handling a corrupt volume was: mkreiserfs ...

3. Then there was the Reiserfs format change ...

4. Ask Chris Mason who it was that explained the nitty-gritty of journaling to
   him (deletion in particular); my recollection is -- Stephen Tweedie!

I never lost data to Ext3, and I started using it with that first
alpha release.  (I did lose data to a laptop IDE chipset, though, a not
infrequent occurrence, along with dodgy RAM.) Over the years, other people
have had issues (orphan handling, etc.)  that were generally fixed with
(a new, improved) e2fsck.  Rarely a case of "my filesystem is gone!"

The Namesys folks have done good work, and continue to do so, and I don't
mean to denigrate them in any way.  Hans just has a habit of shooting
off his mouth, usually with his colleagues cringing and biting their
lips with embarrassment.

Just because SuSE, and later Linus, merged Reiserfs into their trees does not
mean that it was "stable."  Ditto LVM1.

Regards,

	Bill Rugolsky




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]