Ok, I am getting vewy pissed here with a certain Sun U10.  So, 
I set my U30/Solaris 9 box to be a rarpd/tftpd server, first making sure 
the /etc/inted.conf has tftpd enabled and pointing to the right directory:

tftp    dgram   udp6    wait    root    /usr/sbin/in.tftpd      in.tftpd -s
/tftpboot

/etc/ethers has the proper machine name and IP:

# cat /etc/ethers
08:00:20:ff:ec:30 ficus
#

I then download the debian tftpboot.img for sun4u and set it up in
/tftpboot (remember the ip for the U10 is 159.178.77.187.  I used bc to
find out the hex values):

# ln -s debian-3.0-ultra-tftpboot.img 9FB24DBB.SUN4U
#

Then, I start the rarpd in debugging mode to see what is happening and
reboot the U10:

# /usr/sbin/in.rarpd -d -a
/usr/sbin/in.rarpd:[1] device hme0 ethernetaddress 8:0:20:8a:3b:6d
/usr/sbin/in.rarpd:[1] device hme0 address 159.178.77.196
/usr/sbin/in.rarpd:[1] device hme0 subnet mask 255.255.0.0
/usr/sbin/in.rarpd:[3] starting rarp service on device hme0 address
8:0:20:8a:3b:6d
/usr/sbin/in.rarpd:[3] RARP_REQUEST for 8:0:20:ff:ec:30
/usr/sbin/in.rarpd:[3] trying physical netnum 159.178.0.0 mask ffff0000
/usr/sbin/in.rarpd:[3] good lookup, maps to 159.178.77.187
/usr/sbin/in.rarpd:[3] immediate reply sent

and that is where it sits.  On the U10's console, we have:

Sun Ultra 5/10 UPA/PCI (UltraSPARC-IIi 440MHz), No Keyboard
OpenBoot 3.25, 256 MB (50 ns) memory installed, Serial #16772144.
Ethernet address 8:0:20:ff:ec:30, Host ID: 80ffec30.



Bad magic number in disk label
Can't open disk label package
Bad magic number in disk label
Can't open disk label package
Bad magic number in disk label
Can't open disk label package
Boot device: net  File and args:

and that is as far as it goes. I check if tftp is running,

# ps -ef | grep tftp
   nobody 2733 2732 0 10:02:54 ?  0:00 in.tftpd -s /tftpboot
   root 2732 178 0 10:02:54 ?  0:00 in.tftpd -s /tftpboot
   root 2971 2539 0 10:34:01 pts/6 0:00 grep tftp
#

Out of frustration, I read the netbsd document on installing on a sparc
box, http://tinyurl.com/as24k, and try to force the server to map my
client's ethernet address to its IP address:

# arp -s ficus 8:0:20:ff:ec:30
#

And then restarted rarpd, but still have no results. What is going on
here?  Why is it not retrieving the image as provided by tftp?
-- 
Mauricio                                raub-kudria-com
(if you need to email me, use this address =)

In article <d45pu2$1gg0$1@spnode25.nerdc.ufl.edu>,
	Mauricio Tavares <uce@ftc.gov> writes:
> And then restarted rarpd, but still have no results. What is going on
> here?  Why is it not retrieving the image as provided by tftp?

Try running snoop -d hme0 ether 8:0:20:ff:ec:30
on the server and see what the traffic looks like.
You might see some tftp errors being returned,
or maybe not the file(s) you were expecting.

-- 
Andrew Gabriel

Andrew Gabriel wrote:
> In article <d45pu2$1gg0$1@spnode25.nerdc.ufl.edu>,
> 	Mauricio Tavares <uce@ftc.gov> writes:
> 
>>And then restarted rarpd, but still have no results. What is going on
>>here?  Why is it not retrieving the image as provided by tftp?
> 
> 
> Try running snoop -d hme0 ether 8:0:20:ff:ec:30
> on the server and see what the traffic looks like.
> You might see some tftp errors being returned,
> or maybe not the file(s) you were expecting.
> 
	You were right! It is giving back a TFTP Error: access violation 
message. Permissions? The file is world readable

-- 
Mauricio                                raub-kudria-com
(if you need to email me, use this address =)

Mauricio Tavares wrote:
> Andrew Gabriel wrote:
> 
>> In article <d45pu2$1gg0$1@spnode25.nerdc.ufl.edu>,
>>     Mauricio Tavares <uce@ftc.gov> writes:
>>
>>> And then restarted rarpd, but still have no results. What is going on
>>> here?  Why is it not retrieving the image as provided by tftp?
>>
>>
>>
>> Try running snoop -d hme0 ether 8:0:20:ff:ec:30
>> on the server and see what the traffic looks like.
>> You might see some tftp errors being returned,
>> or maybe not the file(s) you were expecting.
>>
>     You were right! It is giving back a TFTP Error: access violation 
> message. Permissions? The file is world readable
> 

     I ran snoop as you suggested and it is giving a TFTP Error: access 
violation message:

# snoop -d hme0 ether 8:0:20:ff:ec:30
Using device /dev/hme (promiscuous mode)
poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu UDP D=41993 S=38680 
LEN=30
ficus.biostat.ufl.edu -> BROADCAST    TFTP Read "9FB24DBB" (octet)
poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu TFTP Error: access 
violation
ficus.biostat.ufl.edu -> BROADCAST    TFTP Read "9FB24DBB" (octet)
poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu TFTP Error: access 
violation
ficus.biostat.ufl.edu -> BROADCAST    TFTP Read "9FB24DBB" (octet)
poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu TFTP Error: access 
violation
ficus.biostat.ufl.edu -> BROADCAST    TFTP Read "9FB24DBB" (octet)
poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu TFTP Error: access 
violation
[...]

So, I googled and found
http://supportforum.sun.com/sunos/index.php?t=msg&goto=2269&rid=0#msg_2269.
Tried their suggestion (chmod 755 on the original file) and restarted 
rarpd in debug  mode (again). Same result. =(

So, I went to another of our Sun boxes (An U1.  Gasp!) in the same room 
and subnet:

mauricio@papaya-16>tftp poisonivy
tftp> get 9FB24DBB.SUN4U
Received 4626800 bytes in 11.1 seconds
tftp> mauricio@papaya-17>

=(


-- 
Mauricio                                raub-kudria-com
(if you need to email me, use this address =)

Mauricio Tavares <uce@ftc.gov> wrote:
>>>> And then restarted rarpd, but still have no results. What is going on
>>>> here?  Why is it not retrieving the image as provided by tftp?

Your first message said:

# ln -s debian-3.0-ultra-tftpboot.img 9FB24DBB.SUN4U

>      I ran snoop as you suggested and it is giving a TFTP Error: access 
> violation message:

> # snoop -d hme0 ether 8:0:20:ff:ec:30
> Using device /dev/hme (promiscuous mode)
> poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu UDP D=41993 S=38680 
> LEN=30
> ficus.biostat.ufl.edu -> BROADCAST    TFTP Read "9FB24DBB" (octet)
> poisonivy.biostat.ufl.edu -> ficus.biostat.ufl.edu TFTP Error: access 
> violation

But there's the the file it's looking for.  You should also link that
file to 9FB24DBB (no .SUN4U extension).

> mauricio@papaya-16>tftp poisonivy
> tftp> get 9FB24DBB.SUN4U

Note the filename...

> Received 4626800 bytes in 11.1 seconds
> tftp> mauricio@papaya-17>

> =(

-- 
Darren Dunham                                           ddunham@taos.com
Senior Technical Consultant         TAOS            http://www.taos.com/
Got some Dr Pepper?                           San Francisco, CA bay area
         < This line left intentionally blank to confuse you. >

Darren Dunham wrote:
> 
> But there's the the file it's looking for.  You should also link that
> file to 9FB24DBB (no .SUN4U extension).
> 
> 
>>mauricio@papaya-16>tftp poisonivy
>>tftp> get 9FB24DBB.SUN4U
> 
> 
> Note the filename...
>
	You are right! So, the Ultras do not need the .SUN4U exntension? That's 
good to know; I guess I spent too much time with my SS20 at home.

Thanks!


-- 
Mauricio                                raub-kudria-com
(if you need to email me, use this address =)

Mauricio Tavares <uce@ftc.gov> wrote:
> 	You are right! So, the Ultras do not need the .SUN4U exntension? That's 
> good to know; I guess I spent too much time with my SS20 at home.

Probably true.  As a rule, I always make both links (with and without
the architecture suffix) because I can't remember which systems need
it and which don't.  I think the 'add_install_client' script for
jumpstarting would do the same thing.

Also, the old netboot sequence uses RARP, tftp, bootparams (not BootP)
and NFS.  I recall that if there were an error in there during the boot
and you *didn't* see any messages on the screen, then it was TFTP that
was screwing up.  The others would eventually give some sort of error.

-- 
Darren Dunham                                           ddunham@taos.com
Senior Technical Consultant         TAOS            http://www.taos.com/
Got some Dr Pepper?                           San Francisco, CA bay area
         < This line left intentionally blank to confuse you. >