netboot issues, 8.0, mfsroot mount failure

[Charles Sprickman]

On Tue, 16 Feb 2010, Garrett Cooper wrote:

> On Tue, Feb 16, 2010 at 5:28 PM, Charles Sprickman <spork at bway.net> wrote:
>> Howdy,
>>
>> I'm having some problems getting 8.0 to install over the network.  I've got
>> my dhcp, tftp and nfs server working well, and I've tested all three
>> services from this host before attempting to boot over the network.
>>
>> pxeboot seems to work, and I see it get loaded via tftp.  The kernel boots,
>> and parses the options in loader.conf that exist in my nfs-exported 8.0 DVD
>> fileset:
>>
>> [root at archive /home/spork/tmp]# cat
>> /usr/local/netboot/freebsd8/boot/loader.conf
>> mfsroot_load="YES"
>> mfsroot_type="mfs_root"
>> mfsroot_name="/boot/mfsroot"
>> boot_multicons="YES"
>> boot_serial="YES"
>> console="comconsole,vidconsole"
>> vfs.root.mountfrom="ufs:/dev/md0c"
>>
>> I see the kernel does find mfsroot and attaches it:
>>
>> md0: Preloaded image </boot/mfsroot> 4423680 bytes at 0xc0f6dfe0
>>
>> But then when it's ready to mount the root filesystem, I get this:
>>
>> SMP: AP CPU #1 Launched!
>> Trying to mount root from ufs:/dev/md0c
>> ROOT MOUNT ERROR:
>>
>> If you have invalid mount options, reboot, and first try the following from
>> the loader prompt:
>>
>>     set vfs.root.mountfrom.options=rw
>>
>> and then remove invalid mount options from /etc/fstab.
>>
>> It doesn't really state what the error is.  It's hinting that it's
>> read-only, but that seems odd.  Even if it couldn't mount r/w, shouldn't it
>> just drop to single-user at this point?
>>
>> Next it tries nfs:
>>
>> Trying to mount root from nfs:
>> NFS ROOT: 192.168.1.111:/usr/local/netboot/freebsd8
>> em0: link state changed to UP
>>
>> And there it sits.  Remotely I can't do anything.  If I'm local, I can
>> ctrl-alt-del a few times and then about a minute later it does an orderly
>> restart.
>>
>> I'm not aware of a good way to snoop on nfs traffic, but tcpdump shows nfs
>> traffic between the two hosts, which appears to be the client stat-ing a
>> file or directory.  tcpdump also shows some checksum errors, but I recall a
>> few threads here mentioning that on Intel cards that generally is not a
>> cause for concern.
>>
>>> From another host, I have no issues mounting that nfs filesystem r/w:
>>
>> root at h10[/home/spork]# mount_nfs 192.168.1.111:/usr/local/netboot/freebsd8
>> /mntroot at h10[/home/spork]# ls /mnt/
>> .cshrc          HARDWARE.TXT    boot.catalog    media           sbin
>> .profile        README.HTM      cdrom.inf       mnt             stand
>> 8.0-RELEASE     README.TXT      dev             packages        sys
>> COPYRIGHT       RELNOTES.HTM    docbook.css     proc            tmp
>> ERRATA.HTM      RELNOTES.TXT    etc             rescue          usr
>> ERRATA.TXT      bin             lib             root            var
>> HARDWARE.HTM    boot            libexec         rr_moved
>> root at h10[/home/spork]# touch /mnt/foo
>> root at h10[/home/spork]# rm /mnt/foo
>> root at h10[/home/spork]# umount /mnt
>>
>> Any ideas?  I've got about a dozen remote boxes to upgrade, so I want to
>> totally nail down this procedure.  I've been putting off learning this for a
>> few years, and now I've got an actual need for it.
>
>    I'll be in your shoes in a little bit... I ran into some issues
> when I last tried with NFSv3 on a Solaris server so we'll see how
> things go in the second go-around [with a FreeBSD nfs rootfs server],

In my case the server is FreeBSD (7.2).  It's running with default nfsd 
flags, so I suppose it's offering both tcp and udp.  No idea what version. 
It seems to work enough for the loader to fetch the loader files and 
kernel...

> but 7.x netbooted and 8.x didn't when I tried last; the 7.x images
> have some secret sauce fixes for PXE booting -- the ones I know of are
> as follows (apply to loader.conf as you feel fit):
>
> vfs.root.mountfrom="nfs"
> boot.nfsroot.path="/absolute/path/to/netboot/dir"
> boot.nfsroot.server="nfs-server-ip-addr"

Is this documented somewhere?

>    There were also some code changes made to `fix' netbooting with
> pxeloader, but I'm not sure if they're applicable or needed in 8.x,
> and I'm not sure what the actual changes are TBH...

Ugh.  With all these variables AND the general nuttiness of PC hardware I 
see many reboot cycles in my future.

Charles


> Cheers,
> -Garrett
>