Re: Openattr issues/clarifications

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

From: Spencer Shepler (shepler@eng.sun.com)
Date: 06/18/01-02:30:46 PM Z


Date: Mon, 18 Jun 2001 14:30:46 -0500
From: Spencer Shepler <shepler@eng.sun.com>
Subject: Re: Openattr issues/clarifications
Message-ID: <20010618143045.A516@dhcp-aus08-229.eng.sun.com>


The protocol is written in a way that allows named attributes
on all file system objects.

On Mon, Khan, Saadia wrote:
> Are named attributes only supported for files or can directories 
> have named attributes too?
> 
> thanks,
> Saadia
> 
> -----Original Message-----
> From: Spencer Shepler [mailto:shepler@eng.sun.com]
> Sent: Monday, June 18, 2001 7:09 AM
> To: 'nfsv4-wg@sunroof.eng.sun.com'
> Subject: Re: Openattr issues/clarifications
> 
> 
> On Sun, Noveck, Dave wrote:
> > 
> > I think the simplest way of accommodating multiple styles of
> > implementation of this feature is to add a boolean parameter to
> > OPENATTR.  If FALSE, OPENATTR would fail if there were no
> > named attributes associated with the file.  This would be
> > open for interrogate.  If TRUE, it would not fail if there
> > were not attributes (or no such directroy although that is
> > not something that is explicitly visible in the protocol).
> > That would open to modify and it would fail on a read-only
> > fs.  Creation per se of the named attribute directory would
> > not be called out as a protocol-visible event but it would
> > allow an implementation to avoid superfluous creations when
> > only interrogations are required. 
> > 
> 
> This seems like a reasonable protocol modification.
> 
> Anyone else have an opinion?
> 
> Spencer

-- 

- Spencer -


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 : 03/04/05-01:48:50 AM Z CST