Dear Group,

I have install and configure Jumpstart for Solaris 9 4/04 on an sparc 
hardware with solaris 8 OS.

The Server Config is:
SERVER:/export/jumpstart
SERVER:/export/jump_5.9

share:
-               /export/jumpstart   ro,anon=0   ""
-               /export/jump_5.9   ro,anon=0   ""

The Client add_script:
add_install_client -c SERVER:/export/jumpstart/ -s 
SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u

The Server accepted this line:
Adding Ethernet number for nc11 to /etc/ethers
making /tftpboot
updating /etc/bootparams
copying inetboot to /tftpboot

All needed processes (in.rarpd and inetd) are running.

The demon in.tftpd is started when the client boot over the net.

After "boot net - install" on the client, he find and boot the kernel, 
then he makes a reset and boot from
the harddisk.

Can any help me?

Juergen Berberich wrote:
> Dear Group,
> 
> I have install and configure Jumpstart for Solaris 9 4/04 on an sparc 
> hardware with solaris 8 OS.
> 
> The Server Config is:
> SERVER:/export/jumpstart
> SERVER:/export/jump_5.9
> 
> share:
> -               /export/jumpstart   ro,anon=0   ""
> -               /export/jump_5.9   ro,anon=0   ""
> 
> The Client add_script:
> add_install_client -c SERVER:/export/jumpstart/ -s 
> SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u
> 
> The Server accepted this line:
> Adding Ethernet number for nc11 to /etc/ethers
> making /tftpboot
> updating /etc/bootparams
> copying inetboot to /tftpboot
> 
> All needed processes (in.rarpd and inetd) are running.
> 
> The demon in.tftpd is started when the client boot over the net.
> 
> After "boot net - install" on the client, he find and boot the kernel, 
> then he makes a reset and boot from
> the harddisk.
> 
> Can any help me?
> 
> 
Did you configure sysidcfd, rules and profiles?

Juergen Berberich <trash-007@gmx.li> wrote:
> Dear Group,

> I have install and configure Jumpstart for Solaris 9 4/04 on an sparc 
> hardware with solaris 8 OS.

What hardware?

> The Server Config is:
> SERVER:/export/jumpstart
> SERVER:/export/jump_5.9

> share:
> -               /export/jumpstart   ro,anon=0   ""
> -               /export/jump_5.9   ro,anon=0   ""

> The Client add_script:
> add_install_client -c SERVER:/export/jumpstart/ -s 
> SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u

You almost never want to use -s.  Unless you're running this on a
boot-only server, the script will figure out the right path.  However if
this were the problem, you'd probably get a nice message like "cannot
mount /cdrom", not a reset.

> The Server accepted this line:
> Adding Ethernet number for nc11 to /etc/ethers
> making /tftpboot
> updating /etc/bootparams
> copying inetboot to /tftpboot

> All needed processes (in.rarpd and inetd) are running.
> The demon in.tftpd is started when the client boot over the net.

You wouldn't get as far as the kernel if they weren't.

> After "boot net - install" on the client, he find and boot the kernel, 
> then he makes a reset and boot from
> the harddisk.

Posting the console output up to the reset might help.

-- 
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. >

Zbigniew Szmigiero <zbychleech@aster.pl> wrote:
>> After "boot net - install" on the client, he find and boot the kernel, 
>> then he makes a reset and boot from
>> the harddisk.
>> 
>> 
> Did you configure sysidcfd, rules and profiles?

None of those should be necessary.  The client should just drop into an
interactive install if they're missing, not reset and reboot.

-- 
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. >

Thanks for answers!

Darren Dunham wrote:
> Juergen Berberich <trash-007@gmx.li> wrote:
> 
>>Dear Group,
> 
> 
>>I have install and configure Jumpstart for Solaris 9 4/04 on an sparc 
>>hardware with solaris 8 OS.
> 
> 
> What hardware?
> 
> 
>>The Server Config is:
>>SERVER:/export/jumpstart
>>SERVER:/export/jump_5.9
> 
> 
>>share:
>>-               /export/jumpstart   ro,anon=0   ""
>>-               /export/jump_5.9   ro,anon=0   ""
> 
> 
>>The Client add_script:
>>add_install_client -c SERVER:/export/jumpstart/ -s 
>>SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u
> 
> 
> You almost never want to use -s.  Unless you're running this on a
> boot-only server, the script will figure out the right path.  However if
> this were the problem, you'd probably get a nice message like "cannot
> mount /cdrom", not a reset.

The CD-Image and the Jumpstart-Config is not on the same path.
I must defind -s. If i only boot the client with 'boot net" then the Client
crash always.

> 
> 
>>The Server accepted this line:
>>Adding Ethernet number for nc11 to /etc/ethers
>>making /tftpboot
>>updating /etc/bootparams
>>copying inetboot to /tftpboot
> 
> 
>>All needed processes (in.rarpd and inetd) are running.
>>The demon in.tftpd is started when the client boot over the net.
> 
> 
> You wouldn't get as far as the kernel if they weren't.

In this machine was always install a jumpstart installserver vor solaris 8!

> 
> 
>>After "boot net - install" on the client, he find and boot the kernel, 
>>then he makes a reset and boot from
>>the harddisk.
> 
> 
> Posting the console output up to the reset might help.
> 

Regrettably can i read the output. The client restart very fast and
i cant read the messages.

Juergen Berberich <trash-007@gmx.li> wrote:
> Thanks for answers!

> Darren Dunham wrote:
>> Juergen Berberich <trash-007@gmx.li> wrote:

>>>The Client add_script:
>>>add_install_client -c SERVER:/export/jumpstart/ -s 
>>>SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u
>> 
>> 
>> You almost never want to use -s.  Unless you're running this on a
>> boot-only server, the script will figure out the right path.  However if
>> this were the problem, you'd probably get a nice message like "cannot
>> mount /cdrom", not a reset.

> The CD-Image and the Jumpstart-Config is not on the same path.
> I must defind -s. If i only boot the client with 'boot net" then the Client
> crash always.

What happens if you do the following (no -s).  Do you get an error?

# cd /export/jump_5.9/Solaris_9/Tools
# ./add_install_client nc11 sun4u

>>>All needed processes (in.rarpd and inetd) are running.
>>>The demon in.tftpd is started when the client boot over the net.
>> 
>> 
>> You wouldn't get as far as the kernel if they weren't.

> In this machine was always install a jumpstart installserver vor solaris 8!

I don't know what that sentence means.

>>>After "boot net - install" on the client, he find and boot the kernel, 
>>>then he makes a reset and boot from
>>>the harddisk.
>> 
>> 
>> Posting the console output up to the reset might help.
>> 

> Regrettably can i read the output. The client restart very fast and
> i cant read the messages.

Using a serial console makes that easier.  You can capture the output.

-- 
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. >

my 2 cents:

1. as suggested by others: use a serial console

2. start

	snoop <ip_adress_of_the_machine_to_install>

on the installation server *before* doing the boot net on the machine 
to install and see what's going on

 From my experience I would think you're using the wrong kernel. Check 
the link in /tftpboot on the installation server for the machine to install.

It's also worth a try to boot the machine from the CDROM that you used 
to create the image on the installation server.

regards

Bernd




Juergen Berberich wrote:
> Thanks for answers!
> 
> Darren Dunham wrote:
> 
>> Juergen Berberich <trash-007@gmx.li> wrote:
>>
>>> Dear Group,
>>
>>
>>
>>> I have install and configure Jumpstart for Solaris 9 4/04 on an sparc 
>>> hardware with solaris 8 OS.
>>
>>
>>
>> What hardware?
>>
>>
>>> The Server Config is:
>>> SERVER:/export/jumpstart
>>> SERVER:/export/jump_5.9
>>
>>
>>
>>> share:
>>> -               /export/jumpstart   ro,anon=0   ""
>>> -               /export/jump_5.9   ro,anon=0   ""
>>
>>
>>
>>> The Client add_script:
>>> add_install_client -c SERVER:/export/jumpstart/ -s 
>>> SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u
>>
>>
>>
>> You almost never want to use -s.  Unless you're running this on a
>> boot-only server, the script will figure out the right path.  However if
>> this were the problem, you'd probably get a nice message like "cannot
>> mount /cdrom", not a reset.
> 
> 
> The CD-Image and the Jumpstart-Config is not on the same path.
> I must defind -s. If i only boot the client with 'boot net" then the Client
> crash always.
> 
>>
>>
>>> The Server accepted this line:
>>> Adding Ethernet number for nc11 to /etc/ethers
>>> making /tftpboot
>>> updating /etc/bootparams
>>> copying inetboot to /tftpboot
>>
>>
>>
>>> All needed processes (in.rarpd and inetd) are running.
>>> The demon in.tftpd is started when the client boot over the net.
>>
>>

>>
>> You wouldn't get as far as the kernel if they weren't.
> 
> 
> In this machine was always install a jumpstart installserver vor solaris 8!
> 
>>
>>
>>> After "boot net - install" on the client, he find and boot the 
>>> kernel, then he makes a reset and boot from
>>> the harddisk.
>>
>>
>>
>> Posting the console output up to the reset might help.
>>
> 
> Regrettably can i read the output. The client restart very fast and
> i cant read the messages.

In article <ciue3c$fba$1@lnx107.hrz.tu-darmstadt.de>, Juergen Berberich <trash-007@gmx.li> writes:
> Dear Group,
> 
> I have install and configure Jumpstart for Solaris 9 4/04 on an sparc 
> hardware with solaris 8 OS.
> 
> The Server Config is:
> SERVER:/export/jumpstart
> SERVER:/export/jump_5.9
> 
> share:
> -               /export/jumpstart   ro,anon=0   ""
> -               /export/jump_5.9   ro,anon=0   ""
> 
> The Client add_script:
> add_install_client -c SERVER:/export/jumpstart/ -s 
> SERVER:/export/jump_5.9/ -e 8:0:20:a9:29:fd nc11 sun4u
> 
> The Server accepted this line:
> Adding Ethernet number for nc11 to /etc/ethers
> making /tftpboot
> updating /etc/bootparams
> copying inetboot to /tftpboot
> 
> All needed processes (in.rarpd and inetd) are running.
> 
> The demon in.tftpd is started when the client boot over the net.
> 
> After "boot net - install" on the client, he find and boot the kernel, 
> then he makes a reset and boot from
> the harddisk.
> 
> Can any help me?
> 
> 

Your boot image (on the boot server) seems to be corrupt.
The boot server is found by the RARP broadcast, and can be
different from the install server. Maybe there are even two
bootservers responding, with unpredictable results...

Run "snoop nc11" in the subnet where your client boots.

Boot your client with
boot net -vs
Does it report the expected boot server?
Does it display Solaris version and Copyright?
Does it come up into single-user?

If the client crashes too fast, you should use a serial console
(connect another Sparc Station's serial port B with a null modem
cable, and run "tip hardwire").
Or interrupt the boot process by Stop-A just before the crash.

-- 
Michael Tosch
IT Specialist
HP Managed Services
Technology Solutions Group
Hewlett-Packard GmbH
Phone: +49 2407 575 313
Mail: michael.tosch:hp.com