On Fri, 9 Nov 2001, Vladimir V. Saveliev wrote:
>>>   FS Tools   : reiserfsprogs-3.x.0j
>>
>> Get latest tools from
>> ftp://ftp.namesys.com/pub/reiserfsprogs/pre/reiserfsprogs-3.x.0k-pre9.tar.gz
>
> Please stay with 3.x.0j.

  I already made that decision, when 3.x.0k reiserfsck reported many errors
about "wrong sd_block", while 3.x.0j did not.  Mailing list archives
indicates it was a result of the pre series, being, after all, a test
release.  This is not a problem for me.  Sometimes the bleeding edge stuff
is worth running, and sometimes it draws blood.  :)

On Fri, 9 Nov 2001, Vladimir V. Saveliev wrote:
> I think that you should consider using 2. 4. At least reiserfs seems to
> get more testing last time.

  While 2.4 has some appealing features, we are not touching it on
mission-critical systems until the VM issue settles down.  So far, testing
of ReiserFS 3.5.34 on Linux 2.2.19 has been very promising.  Significantly
faster than ext2 for many operations, and the online resize worked
flawlessly.  Impressive.

On Fri, 9 Nov 2001, Vladimir V. Saveliev wrote:
> It is probably possible to break 'r5' and generate names having the same
> value of r5.  But, I do not think that it will ever fail if you do not do
> that intentionally.

  Well, we certainly will not do it intentionally, so I think we will be
safe.

> reiserfs should not scramble filesystem when number of hash collisions reaches
> its limit.
> It should return -125 in 2.2 and -EBUSY in 2.4

  That is fine.  Graceful handling of error conditions is all I ask.  :)

> Thanks,

  Thank *you*!

-- 
Ben Scott           | Net Technologies, Inc.  | 978-462-8795
Network Engineer    | Salisbury, MA, USA      | 866-NTI-LINUX (684-5468)
[EMAIL PROTECTED] | http://www.ntilinux.com | Fax: 978-499-7839

Reply via email to