Re: fsync() fails under NFS, right?

New Message Reply About this list Date view Thread view Subject view Author view Attachment view

From: Alexy Khrabrov (khrabrov@unagi.cis.upenn.edu)
Date: 09/14/99-12:33:54 PM Z


Date: Tue, 14 Sep 1999 13:33:54 -0400 (EDT)
Message-Id: <199909141733.NAA19975@unagi.cis.upenn.edu>
From: Alexy Khrabrov <khrabrov@unagi.cis.upenn.edu>
Subject: Re: fsync() fails under NFS, right?



Greetings --

I was away from my mail, but want to thank all who
responded.  For us, the issue is very crucial, and
it is interesting that it doesn't look as such to
many NFS folks.  The "educated guesses" about our
model are remarkably correct.

What surpises me is that I get an impression
that most NFS folks assume fsync() is supposed to
work under NFS.  However, a dilettante I am, only
having read Tanenbaum book, I always assumed NFS
is stateless -- it doesn't maintain cursors for
each user, then how can it do fsync()??  Flush all
I/O on the file?

Thus, I'd like to hear what do NFS people say
think -- fsync() must work under NFS, if it
doesn't, it's a particular client bug, a correct
server allows fsync() to work, and there's no
theoretical problems preventing NFS model from
having totally operable fsync().  Is that right?

-- 
Cheers,
Alexy Khrabrov -- www.suffix.com -- Segmentation f%^(& 


New Message Reply About this list Date view Thread view Subject view Author view Attachment view

This archive was generated by hypermail 2.1.2 : 03/04/05-01:47:34 AM Z CST