From fnaumann@mail.cs.uni-magdeburg.de Thu Jan 8 13:50:54 2004 Message-ID: <1073566032.3ffd5150759fb@imp4-l.free.fr> Date: Thu, 8 Jan 2004 13:47:12 +0100 From: Xavier Joubert To: mint@fishpool.com Subject: Re: [MiNT] MiNT, CT60 and power off MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit User-Agent: Internet Messaging Program (IMP) 3.2.1 Delivered-To: mint@fishpool.com Delivered-To: mint@lists.fishpool.fi X-ecartis-version: Ecartis v1.0.0 Sender: mint-bounce@lists.fishpool.fi Errors-to: mint-bounce@lists.fishpool.fi X-original-sender: xavier.joubert@free.fr Precedence: bulk List-help: List-unsubscribe: List-ID: X-List-ID: Hi all ! Sorry to send this twice, Konrad. I forgot to update "to:" field again... Selon Konrad Kokoszkiewicz : > So if there is a problem, it seems rather related to the hardware, than to > MiNT. Did it occur once, or you experience it all the time, by the way? I think this problem is hardware, too. Here is my theory : There is one big difference between rebooting, or shuting down ARAnyM and powering off the CT60 : the internal hard disk is powered off too. If its write cache is enabled, it is powered off before it has a chance to write to the disk what sync did some milliseconds before... I already explained this here some days ago... But it looks like there were too many post in this list at that time. The problem should not appear if the disk drive is powered by another PSU (SCSI disk, internal IDE connected to another PSU than the Falcon/CT60 (??),...). The solution is to issue a command to instruct all disk drives to write their cache, and then wait for a confirmation before powering off. I don't know the technical details for this though. This should be done in the sync code, in fact, and not in the poweroff one. Another (bad) solution would be to add a small wait (5 seconds ?) before powering off and hope that this will allow each disk to sync its cache. We can consider that the time we take to move our hand to the power switch is enough : the problem did not appear before... I hope this helps. Best regards, Xavier