Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Sun, 17 Feb 2019 23:51:51 +0000
From:   "Richard W.M. Jones" <rjones@redhat.com>
To:     Pavel Machek <pavel@ucw.cz>
Cc:     smcdowell@cloudbd.io, kernel list <linux-kernel@vger.kernel.org>,
        Andrew Morton <akpm@osdl.org>
Subject: Re: nbd, nbdkit, loopback mounts and memory management
Message-ID: <20190217235151.GZ12500@redhat.com>
References: <20190215191953.GB17897@amd>
 <20190217084458.GY12500@redhat.com>
 <20190217231514.GA10675@amd>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20190217231514.GA10675@amd>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Feb 18, 2019 at 12:15:14AM +0100, Pavel Machek wrote:
> But Shaun reported it happened somehow often for them, so he might
> have a practical test case... better than my theories :-).

Yes, certainly not saying this isn't a problem.

I think the good news is the fix seems quite easy, ie. to add mlockall
and adjust the OOM killer score, as is done currently in the client:

https://github.com/NetworkBlockDevice/nbd/blob/3969c3f81a11a483f267a55ed6665d260dc9e1d2/nbd-client.c#L867-L885
https://github.com/NetworkBlockDevice/nbd/blob/3969c3f81a11a483f267a55ed6665d260dc9e1d2/nbd-client.c#L1219

For now I have added a note in the TODO file to follow up in case we
get a test case or reports of a problem:

https://github.com/libguestfs/nbdkit/commit/72e0afe2e280d895f68941677fafa559ddc3bb0d

Thanks,

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
libguestfs lets you edit virtual machines.  Supports shell scripting,
bindings from many languages.  http://libguestfs.org