From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boris Derzhavets Subject: Re: xen.git branch reorg Date: Sat, 25 Apr 2009 01:58:26 -0700 (PDT) Message-ID: <427767.25883.qm@web56102.mail.re3.yahoo.com> References: <20090425081944.GX24960@edu.joroinen.fi> Reply-To: bderzhavets@yahoo.com Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0682346467==" Return-path: In-Reply-To: <20090425081944.GX24960@edu.joroinen.fi> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: =?iso-8859-1?Q?Pasi_K=E4rkk=E4inen?= Cc: Jeremy Fitzhardinge , Xen-devel , Christophe Saout , Ian Campbell List-Id: xen-devel@lists.xenproject.org --===============0682346467== Content-Type: multipart/alternative; boundary="0-1562974501-1240649906=:25883" --0-1562974501-1240649906=:25883 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Pasi, The most recent boards have integrated Gigabit Ethernet Adapters on PCI-E as usual. My preference for Linux=A0 is Marvel Yukon 8056 PCI-E. It's customers's decision what board to purchase and what OS to install.= =20 I am just supposed to fix bug , either left customer on his own. Boris --- On Sat, 4/25/09, Pasi K=E4rkk=E4inen wrote: From: Pasi K=E4rkk=E4inen Subject: Re: [Xen-devel] xen.git branch reorg To: "Boris Derzhavets" Cc: "Jeremy Fitzhardinge" , "Christophe Saout" , "Xen-devel" , "Ian Campbell" Date: Saturday, April 25, 2009, 4:19 AM On Fri, Apr 24, 2009 at 11:55:19PM -0700, Boris Derzhavets wrote: > Chris, >=20 > I have to apologize. Turning off segmentation offloading may work as well= . > Just now it fixed VNC issue on bare metal - CentOS 5.2 with Atansic Gigabit > Ethernet driver (ASUS P5KR) . I gonna try it at DomUs at my earliest convenience. > Have you tried another NICs? It could be a bug in the driver for that NIC.. -- Pasi =20 > Boris. >=20 > --- On Fri, 4/24/09, Christophe Saout wrote: > From: Christophe Saout > Subject: Re: [Xen-devel] xen.git branch reorg > To: "Jeremy Fitzhardinge" > Cc: bderzhavets@yahoo.com, "Xen-devel" , "Ian Campbell" > Date: Friday, April 24, 2009, 6:39 PM >=20 > Hi Jeremy, >=20 > > > In meantime time i see, that 2.6.30-rc3&rc2&rc1-tip are > affected. > > > Solution is the same as on Solaris xVM Linux DomUs about one > > > year ago - is to disable checksum (nothing else) offloading at > Linux=20 > > > DomUs ( CentOS 5.3, Ubuntu 9.04) > > > > > > /usr/local/sbin/ethtool -K etho tx off > > > > OK, that's a good lead. >=20 > Yes, I've been seeing this too (and meant to investigate it before > claiming there's abug) and I can confirm that turning off segmentation > offloading "cures" the problem here too. >=20 > Now the tcpdump on Dom0 looks interesting. It repeatedly sees a packet > with 2880 byte from DomU coming in, which is then dropped and ICMP > "fragmentation needed" sent back, the DomU resends a 1440 byte packet > (after some delay), which then goes through, but then the next one is a > 2880 byte one again, and so on. >=20 > FYI: My Dom0 is running NAT, in case this is relevant. >=20 > =09Christophe >=20 >=20 =0A=0A=0A --0-1562974501-1240649906=:25883 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
Pasi,

The most recent boards have inte= grated Gigabit Ethernet Adapters on
PCI-E as usual. My preference for Li= nux  is Marvel Yukon 8056 PCI-E.
It's customers's decision what bo= ard to purchase and what OS to install.
I am just supposed to fix bug ,= either left customer on his own.

Boris

--- On Sat, 4/25/0= 9, Pasi K=E4rkk=E4inen <pasik@iki.fi> wrote:
From: Pasi K=E4rkk=E4inen <pasik@iki.fi>
Subject: R= e: [Xen-devel] xen.git branch reorg
To: "Boris Derzhavets" <bderzhave= ts@yahoo.com>
Cc: "Jeremy Fitzhardinge" <jeremy@goop.org>, "Chr= istophe Saout" <christophe@saout.de>, "Xen-devel" <xen-devel@lists= .xensource.com>, "Ian Campbell" <Ian.Campbell@citrix.com>
Date: Saturday, April 25, 2009, 4:19 AM

On Fri, Apr 24, 2009 at 11:5=
5:19PM -0700, Boris Derzhavets wrote:
> Chris,
> 
> I hav=
e to apologize. Turning off segmentation offloading may work as well.
&g=
t; Just now it fixed VNC issue on bare metal - CentOS 5.2 with Atansic
G=
igabit
> Ethernet driver (ASUS P5KR) . I gonna try it at DomUs at my =
earliest
convenience.
>

Have you tried another NICs? It cou=
ld be a bug in the driver for that NIC..

-- Pasi

 
> Bo=
ris.
> 
> --- On Fri, 4/24/09, Christophe Saout <christophe@=
saout.de> wrote:
> From: Christophe Saout <christophe@saout.de&=
gt;
> Subject: Re: [Xen-devel] xen.git branch reorg
> To: "Jere=
my Fitzhardinge" <jeremy@goop.org>
> Cc: bderzhavets@yahoo.com,=
 "Xen-devel"
<xen-devel@lists.xensource.com>, "Ian Campbell"
&l=
t;Ian.Campbell@citrix.com>
> Date: Friday, April 24, 2009,
 6:39 PM
> 
> Hi Jeremy,
> 
> > >  In meantim=
e time i see,  that 2.6.30-rc3&rc2&rc1-tip
are
> affected.=

> > > Solution is the same as on Solaris xVM Linux DomUs about=
 one
> > > year ago -  is  to disable checksum (nothing else) o=
ffloading at
> Linux 
> > > DomUs ( CentOS 5.3, Ubuntu 9.=
04)
> > >
> > > /usr/local/sbin/ethtool -K etho tx =
off
> >
> > OK, that's a good lead.
> 
> Yes,=
 I've been seeing this too (and meant to investigate it before
> clai=
ming there's abug) and I can confirm that turning off segmentation
> =
offloading "cures" the problem here too.
> 
> Now the tcpdump o=
n Dom0 looks interesting.  It repeatedly sees a packet
> with 2880 by=
te from DomU coming in, which is then dropped and ICMP
> "fragmentati=
on needed" sent back, the DomU resends a 1440 byte
packet
>
 (after some delay), which then goes through, but then the next one is a> 2880 byte one again, and so on.
> 
> FYI: My Dom0 is runn=
ing NAT, in case this is relevant.
> 
> =09Christophe
> <=
br>>

=0A=0A --0-1562974501-1240649906=:25883-- --===============0682346467== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel --===============0682346467==--