From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-fsdevel-owner@vger.kernel.org>
Received: from mail01-md.ns.itscom.net ([175.177.155.111]:41031 "EHLO
        mail01-md.ns.itscom.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751210AbdBOB3x (ORCPT
        <rfc822;linux-fsdevel@vger.kernel.org>);
        Tue, 14 Feb 2017 20:29:53 -0500
From: "J. R. Okajima" <hooanon05g@gmail.com>
Subject: Re: overlayfs: allowing for changes to lowerdir
To: Josh England <jjengla@gmail.com>
Cc: linux-fsdevel@vger.kernel.org
In-Reply-To: <CA+ZH+jFBAaRi6VPmf3PBdDZQQMOaT6WUByATqaw1QL5M9+-dxg@mail.gmail.com>
References: <CA+ZH+jFBAaRi6VPmf3PBdDZQQMOaT6WUByATqaw1QL5M9+-dxg@mail.gmail.com>
Date: Wed, 15 Feb 2017 10:29:49 +0900
Message-ID: <8172.1487122189@jrobl>
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

Josh England:
> So, let me pose this somewhat naive question:  Would it be possible to
> simply disable any cacheing performed by the overlay to force all
> reads to go to either the tmpfs upper or the (VFS-cached) NFS lower?
> Would this be enough to accomplish my goal of being able to change the
> lowerdir of an active overlayfs?

For your information, aufs allows by-passing aufs, ie. the modification
on the layers or UDBA (user's direct branch access). Internally aufs
sets the inotify to the cached dir inodes and recieves the notification
when the files on the branches (layers) are modified. Recieving the
notification, aufs marks the cached inode obsolete. Next time when the
access happens, aufs_d_revalidate() checks the mark and forces VFS to to
re-lookup.
You can enable/disable this feature anytime by
"mount -o remount,udba=brabra ..."


J. R. Okajima