Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   "Benjamin Coddington" <bcodding@redhat.com>
To:     "Trond Myklebust" <trondmy@hammerspace.com>
Cc:     neilb@suse.de, linux-nfs@vger.kernel.org, anna.schumaker@netapp.com
Subject: Re: [PATCH] NFSv4: use unique client identifiers in network
 namespaces
Date:   Mon, 28 Feb 2022 10:33:41 -0500
Message-ID: <786D4105-B1D4-47B9-9B0B-166F6C0E6ABD@redhat.com>
In-Reply-To: <88a6445c7d7cde971479cae70ca0eb62f68ff96e.camel@hammerspace.com>
References: <6e05bf321ff50785360e6c339d111db368e7dfda.1644349990.git.bcodding@redhat.com>
 <189aba691b341f64f653817c9cdf018ef34ac257.camel@hammerspace.com>
 <7CDAEA98-3A8D-459A-80FE-82AA58A4EDA2@redhat.com>
 <164600585213.15631.6635814364853064416@noble.neil.brown.name>
 <BA04DD0D-4B05-48AD-9E22-AB7524E737C3@redhat.com>
 <88a6445c7d7cde971479cae70ca0eb62f68ff96e.camel@hammerspace.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 28 Feb 2022, at 10:16, Trond Myklebust wrote:
> On Mon, 2022-02-28 at 09:43 -0500, Benjamin Coddington wrote:
>> On 27 Feb 2022, at 18:50, NeilBrown wrote:
>>> I liked this patch at first, but I no longer think it is a good
>>> idea.
>>>
>>> It is quite possible today for containers to provide sufficient
>>> uniqueness simply by setting a unique hostname before the first NFS
>>> mount.
>>> Quite possibly some containers already do this, and are working
>>> perfectly.
>>>
>>> If we add new randomness, the they will get a different identifier
>>> on
>>> each boot.  This is bad for exactly the same reason that it is bad
>>> for
>>> "net == &init_net".
>>>
>>> i.e.  I believe this patch could cause a regression for working
>>> sites.
>>> The regression may not be immediately obvious, but it may still be
>>> there.
>>> For this reason, I think this patch should be dropped.
>>
>> I agree, Trond please drop this patch.
>>
>>
> Since it was already pushed to linux-next, it will have to be reverted.
>

I'm not seeing it on any branch on linux-next.  I presume no one's pulled
your linux-next branch.  Are you not able to drop it from your linux-next
branch?

Ben