From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264717AbUEKN3j (ORCPT ); Tue, 11 May 2004 09:29:39 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264721AbUEKN3j (ORCPT ); Tue, 11 May 2004 09:29:39 -0400 Received: from stormdbn.stormnet.co.za ([196.22.196.1]:19156 "EHLO stormdbn.stormnet.co.za") by vger.kernel.org with ESMTP id S264717AbUEKN3h (ORCPT ); Tue, 11 May 2004 09:29:37 -0400 Subject: weird clock problem From: Nelis Lamprecht Reply-To: nelis@brabys.co.za To: linux-kernel@vger.kernel.org Content-Type: text/plain Organization: Brabys Holdings Message-Id: <1084282171.8334.46.camel@nelis.brabys.co.za> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 Date: Tue, 11 May 2004 15:29:31 +0200 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, Over the past few weeks I've been having major problems keeping time on my machine with kernel 2.6.5. At first I thought it was a problem with ntpd but as it turns out it's my kernel. The problem became evident while copying vast amounts of data across to my machine. While I was copying data to it via scp my random Xscreensaver kicked in displaying the clock and the first thing I noticed was that the clock was advancing at a rapid rate. At the same time I could not type anything as it would just repeat everything I typed 10 fold. Basically the whole system behaved like it was on steroids while I was copying to it and by the time I had finished copying the clock was 2hrs ahead of time. With kernel 2.6.5 ntpd would work on startup and then die saying no servers could be reached which I assume was because my clock was so far off. I have since downgraded to 2.6.3 and now ntpd is keeping time as it should. Anybody else come across this behaviour ? I'm running Gentoo on an IBM ThinkCentre P4 3ghz. I also have APM enabled in both kernels. Cheers, Nelis ps. Please CC me in on your reply as I am not subscribed.