From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-unionfs-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:38688 "EHLO mx1.redhat.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S2438907AbfFLMcQ (ORCPT <rfc822;linux-unionfs@vger.kernel.org>);
        Wed, 12 Jun 2019 08:32:16 -0400
Date: Wed, 12 Jun 2019 08:32:15 -0400
From: Vivek Goyal <vgoyal@redhat.com>
Subject: Re: [PATCH v2] overlay: allow config override of metacopy/redirect
 defaults
Message-ID: <20190612123215.GA27088@redhat.com>
References: <f5b0bddd-678b-bdd9-6fc7-cc9e5b3211e5@gmail.com>
 <CAOQ4uxjQQcrcpxhtu3kAJvGaK+xd5TfNB=7_UnNciGj990DN6Q@mail.gmail.com>
 <CAJfpegvy-Vfc6AEP7+=VfUtfL4izY8AzgoUdvqP4PHnLDEQhNg@mail.gmail.com>
 <20190610184043.GD25290@redhat.com>
 <20190610184553.GE25290@redhat.com>
 <CAJfpegvrOy3yBpu1AVBFyjdXBNM44k4gSqQ0F2npBG8wH8cUeg@mail.gmail.com>
 <20190611130932.GA28835@redhat.com>
 <cb363beb-9b2e-1d20-ca46-cba7724ec648@redhat.com>
 <20190611214951.GC28835@redhat.com>
 <be9bfd25-ff48-bd9e-25ff-aa2a5f5873ed@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <be9bfd25-ff48-bd9e-25ff-aa2a5f5873ed@gmail.com>
Sender: linux-unionfs-owner@vger.kernel.org
To: Matt Coffin <mcoffin13@gmail.com>
Cc: Daniel Walsh <dwalsh@redhat.com>, Miklos Szeredi <miklos@szeredi.hu>, Amir Goldstein <amir73il@gmail.com>, overlayfs <linux-unionfs@vger.kernel.org>, Nalin Dahyabhai <nalin@redhat.com>
List-ID: <linux-unionfs@vger.kernel.org>

On Tue, Jun 11, 2019 at 03:57:03PM -0600, Matt Coffin wrote:
> This could just be because I don't understand the implications here, but
> wouldn't it be easier, at least for now, to just mount with
> 
> redirect_dir=0,metacopy=0
> 
> in the mount parameters when building images, but allow the user's
> default settings to still take over when just executing a container?

I think that's what docker does by default, isn't it? That is redirect_dir
and metacopy features are disabled by default (until and unless user
decides to enable it).

Dan walsh recently changed podman to enable metacopy feature by default
(which in-turn will enable redirect as well). He wants to make use of
user namespaces with containers and chown the images with metacopy
enabled. We don't have shiftfs upstream yet.

BTW, how slow is image building with naivediff interface.

Vivek

> 
> On 6/11/19 3:49 PM, Vivek Goyal wrote:
> > On Tue, Jun 11, 2019 at 05:44:33PM -0400, Daniel Walsh wrote:
> >> On 6/11/19 9:09 AM, Vivek Goyal wrote:
> >>> On Tue, Jun 11, 2019 at 02:37:34PM +0200, Miklos Szeredi wrote:
> >>>> On Mon, Jun 10, 2019 at 8:45 PM Vivek Goyal <vgoyal@redhat.com> wrote:
> >>>> AFAICS what happens when generating a layer is to start with a clean
> >>>> upper layer, do some operations, then save the contents of the upper
> >>>> layer.  If redirect or metacopy is enabled, then the contents of the
> >>>> upper layer won't be portable.  So need to do something like this:
> >>>>
> >>>> traverse(overlay_dir, upper_dir, target_dir)
> >>>> {
> >>>>     iterate name for entries in $upper_dir {
> >>>>         if ($name is non-directory) {
> >>>>             copy($overlay_dir/$name, $target_dir/$name)
> >>>>         } else if ($name is redirect) {
> >>>>             copy-recursive($overlay_dir/$name, $target_dir/$name)
> >>>>         } else {
> >>>>             copy($overlay_dir/$name, $target_dir/$name)
> >>>>             traverse($overlay_dir/$name, $upper_dir/$name, $target_dir/$name)
> >>>>         }
> >>>>     }
> >>>> }
> >>>>
> >>>> Basically: traverse the *upper layer* but copy files and directories
> >>>> from the *overlay*.  Does that make sense?