Re: [blfs-book] [blfs] branch trunk updated: Clarify NFS-v4 vs NFS-v3 in the nfs-utils page.

On 3/1/26 11:18 PM, Git Owner ([email protected] via blfs-book 
Mailing List) wrote:
> This is an automated email from the git hooks/post-receive script.
>
> git pushed a commit to branch trunk
> in repository blfs.
>
> The following commit(s) were added to refs/heads/trunk by this push:
>       new 28db608742 Clarify NFS-v4 vs NFS-v3 in the nfs-utils page.
> 28db608742 is described below
>
> commit 28db6087422a7a3d116f970415b077b5c1ed0586
> Author: Bruce Dubbs <[email protected]>
> AuthorDate: Sun Mar 1 23:18:01 2026 -0600
>
>      Clarify NFS-v4 vs NFS-v3 in the nfs-utils page.
> ---
>   introduction/welcome/changelog.xml |  4 ++++
>   networking/netprogs/nfs-utils.xml  | 40 +++++++++++++++++++-------------------
>   2 files changed, 24 insertions(+), 20 deletions(-)

[snip]

>         <para>
> -        Even if neither end of the connection supports nfs v4, adding
> +        Even if neither end of the connection supports NFSv4, adding
>           <literal>nfsver=3</literal> is still beneficial because it prevents an
>           error message "NFS: bad mount option value specified: minorversion=1"
>           being logged on every mount.
>         </para>
> +      <para>
> +        Note that NFSv3 is a faster, stateless, and simpler protocol ideal for
> +        legacy systems and high-performance VMware environments. NFSv4 is a
> +        stateful, secure protocol (supporting ACLs and Kerberos) that uses a
> +        single TCP port (2049) for easier firewall traversal.
> +      </para>
>       </note>

I have to ask, was this written by AI? I admittedly haven't kept up with 
the ticket much as I've been busy working on other things for *LFS all 
day, but this is almost completely irrelevant to the book.

For our use case this makes absolutely zero sense. Why do we care 
whether it's a stateful or stateless protocol, and why are we mentioning 
high-performance VMWare environments (e.g. ESXi/vSphere) as being 
relevant for BLFS? Why are we also mentioning firewall traversal?

This also effectively calls NFSv3 a protocol for legacy systems which 
probably undermines what we want to say here, as the underlying reason 
for this commit seems to be not supporting NFSv4. For us, at least from 
my interpretation, we're not supporting NFSv4 because of configuration 
difficulties.

I think something like this might be better: "Note that for the purposes 
of BLFS, we are only covering NFSv3 because it is easier to configure. 
If you depend on some of the more advanced configuration or security 
features of NFSv4, you may wish to review 
https://wiki.debian.org/NFSServerSetup" (or a similar link that provides 
more context relevant to configuration of an NFSv4 server).

- Doug

-- 
http://lists.linuxfromscratch.org/sympa/info/blfs-dev
Unsubscribe: See the above information page