[syslinux] Syslinux 4.04 gpxelinux.0 http performance problem with VMware VMs

[Pasi Kärkkäinen]

On Thu, Sep 13, 2012 at 05:15:57AM -0400, Gene Cumm wrote:
> On Sep 12, 2012 1:01 PM, "Pasi Kärkkäinen" <pasik at iki.fi> wrote:
> >
> > On Tue, May 08, 2012 at 10:21:50PM +0300, Pasi Kärkkäinen wrote:
> > > On Mon, May 07, 2012 at 09:53:35PM -0400, Gene Cumm wrote:
> > > > On Mon, May 7, 2012 at 4:40 PM, Pasi Kärkkäinen <pasik at iki.fi> wrote:
> > > > > Hello,
> > > > >
> > > > > In my testing environment I have two VMs on ESXi 5.0.
> > > > >
> > > > > VM A = dhcp/tftp/PXE/http server, running CentOS 6.2. Syslinux 4.04
> with the included gpxelinux.0.
> > > > > VM B = PXE boot client.
> > > >
> > > > > Any ideas what's causing the bad performance and those continuous
> retransmits?
> > > > > Anything I should try to debug this?
> > > > >
> > > > > On baremetal with Intel NICs I've seen much better gpxelinux.0
> download speeds over http.
> > > >
> > > > Syslinux 4.04 is obsolete and probably not worth debugging.  4.05 is
> > > > current and 4.10-pre20 is a beta with lwIP, a lightweight IP stack,
> > > > eliminating the need for gPXE if you only want HTTP downloads.  I've
> > > > already been testing 4.10-pre* in VMware VMs with success after my
> > > > patch.  I've seen a solid 45-70 MB/s after that patch on VMware VMs.
> > > >
> > >
> > > Hello again,
> > >
> > > When using syslinux 4.10-pre20 pxelinux.0 I'm able to get 3-6 MB/sec
> > > over http with VMXNET3 NIC in the VM.
> > >
> > > Did you have to do any special settings (or patches) to get to 45-70
> MB/s ?
> > >
> >
> > Hello,
> >
> > Still wondering about this.. :)
> 
> Sorry about that.
> 
> Nothing really.  I think either ESXi or a hosted product, at least the same
> subnet, possibly the same machine (likely a NAT vmnet if on a hosted
> system),  probably not vlance(pcnet32), a single large file (likely over
> 150 MB), probably lighttpd on Ubuntu-10.04 and I used cptime.c32, comparing
> its stated time to my watch to ensure it's reasonably correct.
> 
> If you're doing 3-6 MB/s across subnets, especially multiple small files
> and/or across a link that's not at least 1Gbps, I'd probably say that might
> be decent.
> 

Ok, thanks for the reply.

My testing was performed in a single subnet, so no routing involved.
The web server was apache2 running on CentOS 5, and I was transferring 
an image size of around 300 MB.

I think the "poor" performance is because of the big amount of retransmits that
I'm seeing with syslinux pxe. When benchmarking the same setup/VMs from Linux
I get almost wirespeed and no restransmits.. 

Something in syslinux pxe stuff is causing those retransmits/drops..

-- Pasi