From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-1?Q?S=E9bastien_RICCIO?= Subject: Re: Dom0 <> DomU clock sync Date: Mon, 03 Sep 2007 16:37:52 +0200 Message-ID: <46DC1C40.7000609@swisscenter.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Keir Fraser Cc: xen-devel List-Id: xen-devel@lists.xenproject.org Hi again, I just tried to use ntpd in dom0. So far nothing changed. The guests are still a couple of minutes behind the dom0. Dom0 and DomU's are set to independent_wallclock 0. Kernels are compiled from the sources included in Xen 3.1. I don't know where to head next :)) Thanks for your help. --=20 S=E9bastien Riccio SwissCenter / OpenBusiness SA sr@openbusiness.com Keir Fraser a =E9crit : > On 3/9/07 15:07, "S=E9bastien RICCIO" wrote: > > =20 >> Ntpd is not running in dom0. I am updating the dom0 time with ntpdate >> (crontab every hour). >> Should ntpd be running on dom0 ? >> =20 > > So long as ntpdate calls settimeofday(), this should be sufficient to u= pdate > Xen's notion of time too. It seems that isn't happening for some reason. > > Do all guests (including dom0) have independent_wallclock equal to 0? > > What kernel are you using in dom0 (e.g., vendor kernel?). > > Ntpd in dom0 would be an interesting thing to try. That should cause do= m0 to > periodically (every few minutes) push its time down to Xen to be picked= up > by all domUs. But really ntpdate should be doing the right thing... > > In our kernels it should be *impossible* for dom0 and domU time to dive= rge > when independent_wallclock is set to zero everywhere... > > -- Keir > > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel > > > =20