From owner-freebsd-questions@FreeBSD.ORG Tue Jun 5 15:17:03 2012 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0CA3C106566B for ; Tue, 5 Jun 2012 15:17:03 +0000 (UTC) (envelope-from martin.dimitrov@mafiainc.org) Received: from mail0.mafiainc.org (mafiainc.org [89.22.97.29]) by mx1.freebsd.org (Postfix) with ESMTP id C67838FC15 for ; Tue, 5 Jun 2012 15:17:02 +0000 (UTC) Received: from [192.168.1.62] (gate.stanga.net [195.34.122.3]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail0.mafiainc.org (Postfix) with ESMTPSA id 1C6102AE2F6 for ; Tue, 5 Jun 2012 18:08:24 +0300 (EEST) Message-ID: <4FCE20E7.2000202@mafiainc.org> Date: Tue, 05 Jun 2012 18:08:23 +0300 From: Martin Dimitrov User-Agent: Mozilla/5.0 (X11; Linux i686; rv:12.0) Gecko/20120428 Thunderbird/12.0.1 MIME-Version: 1.0 To: freebsd-questions@freebsd.org Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Subject: Clock lagging behind on FreeBSD 9.0-RELEASE under KVM X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 05 Jun 2012 15:17:03 -0000 Hi, I am new to FreeBSD, decided to migrate a web server to FreeBSD. I recently both a VPS that claim to use KVM as a virtualization service, I don't know the details of the real hardware running behind nor what is KVM running on. Anyway I have an issue with clock on my FreeBSD installation that I can't live with. The clock is lagging behind, for example running sleep 30 is really sleeping around 35 seconds not 30. Also seems that NTP is not able to manage with this drift in time. Before posting here I red about similar problems mostly related to VMWare guests, but the solutions suggested are following: set kern.hz=100 or kern.hz=50 (doesn't work for me) set hint.apic.0.disabled=1 (this makes the guest hangs while booting also it discarding the SMP capabilities of the kernel which I assume is not a good idea) set kern.timecounter.hardware TSC (doesn't work for me) Is there any chance I deal with this time drifting issue somehow? If somebody faced such issue and managed it I would be happy to try another possible solution? Alternatively I can switch the provider with other that is using Xen for virtualization, I guess is better, but no guarantee that would not have the same issue. :( Cheers, Martin