Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 20 Apr 2021 14:16:27 -0400
From:   "bfields@fieldses.org" <bfields@fieldses.org>
To:     Trond Myklebust <trondmy@hammerspace.com>
Cc:     "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>,
        "ajmitchell@redhat.com" <ajmitchell@redhat.com>,
        "SteveD@RedHat.com" <SteveD@RedHat.com>,
        "chuck.lever@oracle.com" <chuck.lever@oracle.com>
Subject: Re: [PATCH 0/3] Enable the setting of a kernel module parameter from
 nfs.conf
Message-ID: <20210420181627.GA7297@fieldses.org>
References: <2F7FBCA0-7C8D-41F0-AC39-0C3233772E31@oracle.com>
 <c13f792a-71e8-494f-3156-3ff2ac7a0281@RedHat.com>
 <DAFAF7B1-5C56-4DA7-B7F9-4F544CCDA031@oracle.com>
 <f0525973-a32c-32d4-4ccc-827acaa3c125@RedHat.com>
 <20A43DDA-C08E-4E39-A83C-24E326768ADE@oracle.com>
 <2d7d391802a3984b68aa8b3e7f360b0b6cb733dc.camel@hammerspace.com>
 <20210420171806.GC4017@fieldses.org>
 <be1a2b6beab29b3e40277f5fefd6c49b37c32361.camel@hammerspace.com>
 <20210420174036.GD4017@fieldses.org>
 <85b4ca155d1697a714be88a67c505d287e22be46.camel@hammerspace.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <85b4ca155d1697a714be88a67c505d287e22be46.camel@hammerspace.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
Precedence: bulk

On Tue, Apr 20, 2021 at 05:53:34PM +0000, Trond Myklebust wrote:
> So if the machine-id exists, then maybe we could indeed generate the
> identity using the uuid in that file (although the question remains as
> to why you'd want that?).

I was assuming: When you clone a machine image, either you want the
clone to have the same identity (maybe it's a backup, or you're doing
some sort of migration) or you want it to act like a new machine (say
you've got a base image that you're using to make a bunch of hosts).  In
the latter case you've got to track down everything on the filesystem
that needs to differ between hosts and fix it up.  The fewer of those,
the better.

> However the generated value should then be persisted separately so
> that it can be platform independent.

That'd be OK.

I don't think switching a host between systemd and not systemd-based is
a common case.

If somebody has a really weird case they can always write their own
script.

--b.