From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-fsdevel-owner@vger.kernel.org>
Received: from helmsgagent01.f-secure.com ([193.110.108.21]:35489 "EHLO
        helmsgagent01.f-secure.com" rhost-flags-OK-OK-OK-OK)
        by vger.kernel.org with ESMTP id S1754913AbcLUKPl (ORCPT
        <rfc822;linux-fsdevel@vger.kernel.org>);
        Wed, 21 Dec 2016 05:15:41 -0500
From: Marko Rauhamaa <marko.rauhamaa@f-secure.com>
To: Amir Goldstein <amir73il@gmail.com>
CC: Jan Kara <jack@suse.cz>, Al Viro <viro@zeniv.linux.org.uk>,
        Eric Paris <eparis@redhat.com>,
        <linux-fsdevel@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Subject: Re: [RFC][PATCH 0/2] fsnotify: super block watch
References: <1482247207-4424-1-git-send-email-amir73il@gmail.com>
Date: Wed, 21 Dec 2016 12:15:26 +0200
In-Reply-To: <1482247207-4424-1-git-send-email-amir73il@gmail.com> (Amir
        Goldstein's message of "Tue, 20 Dec 2016 17:20:05 +0200")
Message-ID: <87bmw5ziyp.fsf@drapion.f-secure.com>
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

Amir Goldstein <amir73il@gmail.com>:

> On the Oct 10 posting you asked me about the use case and it was hard
> to explain the use case with only part of the work done.
>
> The issue, which this work sets to solve, is the poor scalability of
> recursive inotify watches.

On my [employer's] part, the fanotify API suffers from leakage through
namespaces.

If you need to monitor files in a filesystem, the current fanotify
FAN_MARK_MOUNT flag does not give you a notification if you execute a
command like

   unshare -m touch xyz

inside the file system.

Amir's patch addresses this increasingly critical issue.


Marko