linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Benny Halevy <bhalevy@panasas.com>
To: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Steve Dickson <steved@redhat.com>,
	NFS list <linux-nfs@vger.kernel.org>,
	pNFS Mailing List <pnfs@linux-nfs.org>,
	Chuck Lever <chuck.lever@oracle.com>
Subject: Re: [PATCH 0/3] nfs-utils: nfsd support for minor version, take 2
Date: Thu, 23 Apr 2009 11:58:28 +0300	[thread overview]
Message-ID: <49F02DB4.50504@panasas.com> (raw)
In-Reply-To: <20090422215458.GL9541@fieldses.org>

On Apr. 23, 2009, 0:54 +0300, "J. Bruce Fields" <bfields@fieldses.org> wrote:
> On Wed, Apr 22, 2009 at 03:06:30PM +0300, Benny Halevy wrote:
>> Following the discussion we had last week
>> (see http://linux-nfs.org/pipermail/pnfs/2009-April/007283.html)
>>
>> I made the following changes:
>> * minorvers4 enabled by default.
>> * No new -n option.
>>
>> This makes minorvers control essentially the same as
>> the major protocol version.  minorvers4 is enabled by default
>> and can be disabled using -N 4.1.
>>
>> On Fedora, /etc/sysconfig/nfs can be changed as follows
>> to configure the service startup script to disable 4.1:
> 
> Why couldn't nfs-utils just respect the kernel's default and make no
> attempt to set the minor version?  A (possibly undocumented) -P 4.1
> option (or some other name) could be used by testers to specify that
> they want 4.1.  After things have settled down a little we'd change the
> kernel's default, and then only -N 4.1 would be needed.

It could, and that was pretty much my original intent, which was
criticized by Steve and others for being overly cautious.  I agree with
Chuck that the distributions using nfs-utils should decide about the
default and that can be done either way.  The difference is for people
upgrading nfs-utils while keeping their old /etc/sysconfig/nfs.

I've no problem with having an option to enable 4.1 (and I don't care
much if it's -n or -P either).  Should a distribution that want this option
enabled by default use an undocumented option?  I'm not sure.
I'd prefer it to be documented if that's the case.

> 
> If I'd known we'd be enabling 4.1 by default in nfs-utils, I wouldn't
> have been so happy about removing the 4.1 config option--how confident
> are we that the 4.1 code to be in 2.6.30 has no security holes?

The more it will be tested and used, the better my confidence will be.
I don't think that disabling it by default will help us find any security
holes before the code will be ready for prime time.  Not without a
full blown QA effort.

Benny

> 
> --b.
> 
>> --- /etc/sysconfig/nfs.orig	2009-04-22 14:57:15.000000000 +0300
>> +++ /etc/sysconfig/nfs	2009-04-22 14:46:52.000000000 +0300
>> @@ -26,6 +26,8 @@
>>  #RPCNFSDARGS="-N 2 -N 3"
>>  # Turn off v4 protocol support
>>  #RPCNFSDARGS="-N 4"
>> +# Turn off v4.1 minorversion support
>> +RPCNFSDARGS="-N 4.1"
>>  # Number of nfs server processes to be started.
>>  # The default is 8. 
>>  #RPCNFSDCOUNT=8
>>
>> The patches in this set are:
>> [PATCH 1/3] utils/nfsd: add support for minorvers4
>> [PATCH 2/3] utils/nfsd: disable minorvers4 via command line
>> [PATCH 3/3] utils/nfsd: enable nfs minorvers4 by default
>>
>> Benny

  reply	other threads:[~2009-04-23  8:58 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-13  8:25 [RFC 0/4] nfs-utils: nfsd support for minor version Benny Halevy
2009-04-13  8:29 ` [PATCH 1/4] utils/nfsd: fix -N optarg error printout Benny Halevy
2009-04-13  8:29 ` [RFC 2/4] utils/nfsd: add support for minorvers4 Benny Halevy
2009-04-13  8:29 ` [RFC 3/4] utils/nfsd: add -n --nfs-version option Benny Halevy
2009-04-13  8:29 ` [RFC 4/4] utils/nfsd: enable/disable minorvers4 via command line Benny Halevy
2009-04-16 17:24 ` [RFC 0/4] nfs-utils: nfsd support for minor version Steve Dickson
     [not found]   ` <49E769C5.6010902-AfCzQyP5zfLQT0dZR+AlfA@public.gmane.org>
2009-04-16 17:59     ` Benny Halevy
2009-04-16 18:13       ` Steve Dickson
     [not found]         ` <49E7753C.4010300-AfCzQyP5zfLQT0dZR+AlfA@public.gmane.org>
2009-04-16 18:23           ` [pnfs] " J. Bruce Fields
2009-04-16 18:37             ` Benny Halevy
2009-04-16 19:01             ` Steve Dickson
     [not found]               ` <49E7809B.2020002-AfCzQyP5zfLQT0dZR+AlfA@public.gmane.org>
2009-04-16 19:18                 ` J. Bruce Fields
2009-04-17 12:35                   ` Steve Dickson
     [not found]                     ` <49E87798.8090308-AfCzQyP5zfLQT0dZR+AlfA@public.gmane.org>
2009-04-17 15:42                       ` J. Bruce Fields
2009-04-17 16:18                       ` Chuck Lever
2009-04-17 16:40                         ` Benny Halevy
2009-04-22 12:06 ` [PATCH 0/3] nfs-utils: nfsd support for minor version, take 2 Benny Halevy
2009-04-22 12:10   ` [PATCH 1/3] utils/nfsd: add support for minorvers4 Benny Halevy
2009-04-22 12:10   ` [PATCH 2/3] utils/nfsd: disable minorvers4 via command line Benny Halevy
2009-04-22 12:10   ` [PATCH 3/3] utils/nfsd: enable nfs minorvers4 by default Benny Halevy
2009-04-22 21:54   ` [PATCH 0/3] nfs-utils: nfsd support for minor version, take 2 J. Bruce Fields
2009-04-23  8:58     ` Benny Halevy [this message]
2009-05-18 14:49   ` Steve Dickson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=49F02DB4.50504@panasas.com \
    --to=bhalevy@panasas.com \
    --cc=bfields@fieldses.org \
    --cc=chuck.lever@oracle.com \
    --cc=linux-nfs@vger.kernel.org \
    --cc=pnfs@linux-nfs.org \
    --cc=steved@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).