From mboxrd@z Thu Jan 1 00:00:00 1970 From: "J. Bruce Fields" Subject: Re: Question: why only 8 security flavors per export?? Date: Fri, 4 Dec 2009 13:10:41 -0500 Message-ID: <20091204181041.GA18525@fieldses.org> References: <4B1934A7.90801@RedHat.com> <68742FB7-C923-4219-909F-F47B992584C6@netapp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Steve Dickson , Linux NFS Mailing list To: Andy Adamson Return-path: Received: from fieldses.org ([174.143.236.118]:54131 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932371AbZLDSJe (ORCPT ); Fri, 4 Dec 2009 13:09:34 -0500 In-Reply-To: <68742FB7-C923-4219-909F-F47B992584C6@netapp.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Fri, Dec 04, 2009 at 12:08:30PM -0500, Andy Adamson wrote: > > >> Hello, >> >> Why are only 8 (SECFLAVOR_COUNT) security flavors allow >> per export when 13 security flavors are supported >> (see flav_map[] in nfs/exports.c)? >> >> I'm trying to create an export that supports all possible >> security flavors and its erroring out with "mountd: more than >> 8 security flavors on an export" > > SPKM3 drafts have expired, and the LIPKEY RFC is based on SPKM3 and was > never implemented. The SPKM3 implementation was used for > experimentation, but never in production anywhere and should probably be > removed from the kernel. So in reality, the list of all possible > security flavors is: > > AUTH_NONE, AUTH_NULL, AUTH_SYS, AUTH_UNIX, krb5, krb5i and krb5p. Yeah. Until there's some practical use for more than 8 simultaneous flavors, I don't think we care about this limit. --b.