[syslinux] memdisk and iso

[upen]

On Sat, Jun 8, 2013 at 2:20 PM, Gene Cumm <gene.cumm at gmail.com> wrote:

> On Sat, Jun 8, 2013 at 3:10 PM, Gene Cumm <gene.cumm at gmail.com> wrote:
> > On Thu, Jun 6, 2013 at 7:02 PM, upen <upendra.gandhi at gmail.com> wrote:
> >> On Thu, Jun 6, 2013 at 5:38 PM, Gene Cumm <gene.cumm at gmail.com> wrote:
> >>> On Jun 6, 2013 5:42 PM, "H. Peter Anvin" <hpa at zytor.com> wrote:
> >>> > On 06/06/2013 02:40 PM, upen wrote:
> >>> > >
> >>> > > Some odd things I observed just now.
> >>> > > 1. Since yesterday, I have been reseting the client box from
> >>> > > Virtualbox VM
> >>> > > menu (Reset option). This always showed me com32 vesamenu.c32
> message.
> >>> > > 2. Some time ago, I decided I power off the client VM(
> Close->poweroff
> >>> > > from
> >>> > > Menu). This time I saw the menu. It didn't show the background jpg
> >>> > > though.
> >>> > > Anyways I chose the menu option for loading ISO. It returned me the
> >>> > > same
> >>> > > menu options, while showing file not found messages at the bottom
> of
> >>> > > the
> >>> > > screen.
> >>> > > 3. I tried reset as in (1), again com32 -vesamenu.c32 message as
> >>> > > stated
> >>> > > earlier.
> >>> > > 4. Power off , power on . No menu this time. Shows the
> >>> > > com32-vesamenu.c32
> >>> > > message as stated earlier.
> >>> > >
> >>> > > No such issues with pxelinux.0 though.
> >>> > >
> >>> > > I am now rebooting TFTP/DHCP server and see if that helps  to any
> >>> > > level.
> >>> > >
> >>> >
> >>> > Sounds like it's losing lots of packets.
> >>> >
> >>> >         -hpa
> >>> >
> >>>
> >>> What PXE OROM is being used?  VirtualBox version?
> >>
> >> Virtualbox version 4.2.12 r84980. AMD64.
> >> 00:00:01.780076   LanBootRom     <string>  =
> >>
> "/opt/VirtualBox/ExtensionPacks/Oracle_VM_VirtualBox_Extension_Pack/PXE-Intel.rom"
> >> (cb=81
> >>
> >> Thanks.
> >>>
> >>> While working on 4.10-pre, I recall cases others saw with
> >>> dropped/intercepted packets and observed a race in a VM when it was
> using
> >>> the wrong lwIP call.  I was working on some verbose debug code that
> should
> >>> help but need to port it up and finish it.
> >
> > Now that I have this straight, I recall some issue in VirtualBox 4.2
> > that required a downgrade to 4.1.
>
> https://www.virtualbox.org/ticket/8981
> https://www.virtualbox.org/ticket/7385 see ~comment 21
>
> Upen, what host OS name and version/release?  Looks to be a Linux
> system.  Did you compile VirtualBox yourself?
>

Good morning,

I am running Virtual Box on Gentoo/amd64.  VB was installed using
precompiled VirtualBox-4.2.12-84980-Linux_amd64.run available for download.
Then, RHEL 5.8 VM was installed and Virtualbox guest Additions were also
installed on it. This VM is doing the TFTP/DHCP and now webservices. ETH0
on RHEL VM is using bridge adapter for public access. While ETH1(TFTP/DHCP
services) on the RHEL VM is connected to VBox's internal network and
assigned a static IP-192.168.0.100.

Another VM which PXE client has network adapter connected to same switch
ETH! of VM server is connected.

 Thanks.