[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <08de868e-d207-8980-1da4-da79eda07560@auristor.com>
Date: Thu, 11 Mar 2021 09:54:47 -0500
From: Jeffrey E Altman <jaltman@...istor.com>
To: "David Howells (dhowells@...hat.com)" <dhowells@...hat.com>,
linux-afs@...ts.infradead.org
Cc: Gaja Sophie Peters <gaja.peters@...h.uni-hamburg.de>,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/2] AFS metadata xattr fixes
On 3/11/2021 9:10 AM, David Howells (dhowells@...hat.com) wrote:
> I wonder if it's better to just hide all the afs.* xattrs from listxattr().
> It would probably be even better to not use xattrs for this, but I'm not
> sure what I would use instead.
I believe that all of the "afs.*" xattrs should be hidden from
listxattr(). Any "afs.*" xattr that is read from an afs inode can be
copied to another filesystem and stored. Attempts to set these values
in an afs volume will fail.
The use of xattrs is intended to be an alternative to the IBM/OpenAFS
pioctls which are accessed by processes such as "fs", "aklog", "tokens",
etc which would know the names of the xattrs and how to use them. Such
tools would not require listxattr() to find them.
Thanks for the consideration.
Jeffrey Altman
View attachment "jaltman.vcf" of type "text/x-vcard" (272 bytes)
Download attachment "smime.p7s" of type "application/pkcs7-signature" (4033 bytes)
Powered by blists - more mailing lists