From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail.windriver.com ([147.11.1.11]) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1TzeQV-0007Sw-AP for openembedded-core@lists.openembedded.org; Mon, 28 Jan 2013 03:26:27 +0100 Received: from ALA-HCA.corp.ad.wrs.com (ala-hca.corp.ad.wrs.com [147.11.189.40]) by mail.windriver.com (8.14.5/8.14.3) with ESMTP id r0S2Ajaf011428 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Sun, 27 Jan 2013 18:10:45 -0800 (PST) Received: from [128.224.163.154] (128.224.163.154) by ALA-HCA.corp.ad.wrs.com (147.11.189.50) with Microsoft SMTP Server (TLS) id 14.2.328.9; Sun, 27 Jan 2013 18:10:45 -0800 Message-ID: <5105DE37.2090309@windriver.com> Date: Mon, 28 Jan 2013 10:11:03 +0800 From: ChenQi User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130106 Thunderbird/17.0.2 MIME-Version: 1.0 To: Richard Purdie References: <680bfa5f6dd8d3a3e77a18240caab373f03b8006.1359014474.git.Qi.Chen@windriver.com> <1359123271.13917.1.camel@ted> In-Reply-To: <1359123271.13917.1.camel@ted> X-Originating-IP: [128.224.163.154] Cc: openembedded-core@lists.openembedded.org, Zhenfeng.Zhao@windriver.com Subject: Re: [PATCH 09/12] initscripts: inherit volatiles to support read-only rootfs X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 28 Jan 2013 02:26:27 -0000 Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 7bit On 01/25/2013 10:14 PM, Richard Purdie wrote: > On Fri, 2013-01-25 at 13:11 +0000, Burton, Ross wrote: >> On 24 January 2013 08:16, wrote: >>> +VOLATILE_DIR = "/var/volatile/lib/urandom" >>> +VOLATILE_LINK = "/var/lib/urandom" >> A better solution would be to not attempt to save and restore a seed >> value that is intended to persist across reboots when it's on a tmpfs. >> Putting the urandom seed into tmpfs is just a waste of time. > I have to wonder why we don't put a union mount tmpfs over /var/lib to > be honest... > > or if we don't want to depend on unionfs, a tmpfs over /var/lib... > > Cheers, > > Richard > > > A union mount tmpfs over /var/lib (and maybe /etc) was the first solution that came into my mind. Then I saw a bug related to unionfs about getcwd() call, and I saw the unionfs was still disabled for live images. So I thought "maybe I should not use unionfs". A tmpfs over /var/lib. Hmm... I thought about that too. The reason that I didn't use this method was that I thought that some directories under /etc might also need to be written to. It turns out that this concern seems unnecessary. Anyway, I'll try out the tmpfs-over-var-lib method and let you know the result. Thanks, Chen Qi