PXE booting issues

[Morgan Reed]

On 6/25/07, Robert Joosten <robert at ml.erje.net> wrote:
> And next-server

My dhcpd.conf has an appropriate next-server directive, all the
configuration (for the network boot) has been duplicated from the
functional DragonFly boot server.

> I have a couple of version 6.2 pxe-clients booting of a 4.11 box without a
> problem.

My FreeBSD boot host is running 6.2-STABLE

> and the kernel config:
> #grep -i bootp /usr/src/sys/i386/conf/GISTER
> options         BOOTP
> options         BOOTP_NFSROOT
> options         BOOTP_NFSV3
> options         BOOTP_COMPAT
> options         BOOTP_WIRED_TO=fxp0

I'm not using the BOOTP options in my kernel, the same image is able
to get its kernel on another system so I'm happy it's not the image
causing this issue.

At Danny's suggestion I've done some sniffing of the virtual interface
and discovered that none of the NFS calls from the client are being
replied to, yet another host on the same subnet can mount the same
filesystem without issue, that said I've just run another trace, and I
do notice that the calls when I mount the filesystem on the other host
are a mixture of V2 and V3 calls whereas the call from the PXE host is
a V2 call (although this is the GETPORT call which is V2 in both
cases).

Thanks for your time.
Morgan