Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 30 Mar 2021 16:43:23 +0800
From:   Yongcheng Yang <yongcheng.yang@gmail.com>
To:     steved@redhat.com
Cc:     Roberto Bergantinos Corpas <rbergant@redhat.com>,
        linux-nfs@vger.kernel.org
Subject: Re: [PATCH] exportfs: make root unexportable
Message-ID: <20210330084323.GA4033@yoyang-pc.usersys.redhat.com>
References: <20210329105435.17431-1-rbergant@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210329105435.17431-1-rbergant@redhat.com>
Precedence: bulk

Hi SteveD,

This is a real problem introduced by my previous update. Sorry for it.

How about also adding this tag:
    Fixes: a9a7728d (exportfs: Deal with path's trailing "/" in unexportfs_parsed())

Thanks,
Yongcheng

On Mon, Mar 29, 2021 at 12:54:35PM +0200, Roberto Bergantinos Corpas wrote:
> If root of the filesystem is exported, it cannot be explicitly
> unexported, since unexportfs_parsed, in order to deal with trailing
> '/', will leave nlen at 0 for root exported case
> 
> Signed-off-by: Roberto Bergantinos Corpas <rbergant@redhat.com>
> ---
>  utils/exportfs/exportfs.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/utils/exportfs/exportfs.c b/utils/exportfs/exportfs.c
> index 9fcae0b3..9b6f4f5a 100644
> --- a/utils/exportfs/exportfs.c
> +++ b/utils/exportfs/exportfs.c
> @@ -372,7 +372,7 @@ unexportfs_parsed(char *hname, char *path, int verbose)
>  	 * so need to deal with it.
>  	*/
>  	size_t nlen = strlen(path);
> -	while (path[nlen - 1] == '/')
> +	while ((path[nlen - 1] == '/') && (nlen > 1))
>  		nlen--;
>  
>  	for (exp = exportlist[htype].p_head; exp; exp = exp->m_next) {
> -- 
> 2.21.0
>