From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263812AbTKSKGM (ORCPT ); Wed, 19 Nov 2003 05:06:12 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263920AbTKSKGM (ORCPT ); Wed, 19 Nov 2003 05:06:12 -0500 Received: from 44ba20135.mrgnhll.ca.charter.com ([68.186.32.135]:34936 "EHLO uzo.telecoma.net") by vger.kernel.org with ESMTP id S263812AbTKSKGL (ORCPT ); Wed, 19 Nov 2003 05:06:11 -0500 Date: Wed, 19 Nov 2003 02:25:42 -0800 From: Firenza To: linux-kernel@vger.kernel.org Subject: time freeze under 2.4.9 Message-ID: <20031119022542.J777@telecoma.net> Mail-Followup-To: linux-kernel@vger.kernel.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hello, Pretext: 4-way server runs on plain vanilla redhat 2.4.9-21 kernel for almost 2 years without any problems. Two weeks ago, applications start to behave "strangely", which is traced back to the fact that time "froze" on that server. # strace sleep 1 ... nanosleep({1, 0}, (never returns) date shows the freezing time, hwclock shows that the hardware clock is still working and no suspicious kernel messages are found in the logs. Discouragingly, the same thing happened to a different server (same kernel, very similar hardware) a couple of days later. Additionally that server is not in use, which makes it less likely that this behaviour is caused by a freshly introduced application. Rebooting fixed the problem, but the same behaviour surfaced again for the first server a couple of hours ago. What circumstances could cause this behaviour? Is there a way to further pinpoint the cause? cheers, Firenza