Very Computer

rarpd-ss981107-2mdk
bootparamd-0.17-5mdk
tftp-server-0.34-1mdk
nfs-utils-1.0.6-1mdk

0:3:bA:8:7D:98  susan

susan   root=wardrobe:/var/solaris9_install/Solaris_9/Tools/Boot \
        install=wardrobe:/var/solaris9_install/Solaris_9/ \
        boottype=:in \
        rootopts=:rsize=32768

**snip**
/var/solaris9_install/ 10.22.0.0/16(ro)

127.0.0.1               localhost
10.22.10.22             wardrobe
**snip**
10.22.10.33             susan

total 472
-rwxr-xr-x  1 root root 152376 Jun 15 16:37 0A160A21*
-rwxr-xr-x  1 root root 152376 Jun 15 16:37
0A160A21.SUNW,SUN-BLADE-100*
-rwxr-xr-x  1 root root 152376 Jun 15 16:37
inetboot.SUNW,SUN-BLADE-100.Solaris_9-1*
-rw-r--r--  1 root root    383 Jun 15 16:37 rm.10.22.10.33
lrwxrwxrwx  1 root root      2 Jun 15 16:37 tftpboot -> .//

When I run the client with "boot net - install -v", it seems to run
fine until it gets to a line describing the CPU:

cpu0: UltraSparc-IIe (portid 0 impl 0x13 ver 0x14 clock 502Mhz)

and then proceeds to sit there forever. During the output, susan does
correctly get the ip address, the hostname, and the domain name. If
someone tells me how to get the bootup output from a log or file, I'll
post it here for you to see.

The install files were originally setup using the setup scripts from
the Solaris 9 CDs on one of the workstations, and then copied over to
the server. The four processes on the server are running, and I have
temporarily taken down the firewall (shorewall).

I have been using documentation from these sites:
http://www.docbert.org/Solaris/Jumpstart/linux.html
http://www.scsy.de/~mas/archiv/000017.html

If it makes a difference, I have the server providing NIS and NFS to
the machines for login and home directories. The /etc/nsswitch.conf
file is setup to serve through NIS, but everything (on the server)
looks to files first.

Thanks in advance for your help - let me know if there is anything
else I can provide to help debug this!

Chris Phillips
Westmont College

http://groups.google.com/groups?th=7df5a88654120b61

msgs 11 and 18-20 may be relevant.

--

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. >

Sami

--
.signature: No such file or directory

Quote:> Have you tried to snoop network traffic while this happens to find
> out what excactly stalls the install?

tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size
96 bytes
18:31:08.315060 rarp who-is susan tell susan
18:31:12.928579 rarp who-is susan tell susan
18:31:12.952701 IP (tos 0x0, ttl 255, id 1, offset 0, flags [DF],
length: 104) susan.biff > 255.255.255.255.sunrpc: UDP, length: 76
18:31:12.954722 arp who-has wardrobe (Broadcast) tell susan
18:31:12.956485 arp who-has wardrobe (Broadcast) tell susan
18:31:26.264281 rarp who-is susan tell susan
18:31:26.372478 arp who-has susan (Broadcast) tell susan
18:31:26.373633 IP (tos 0x0, ttl   1, id 5961, offset 0, flags [DF],
length: 124) susan.1023 > 255.255.255.255.sunrpc: UDP, length: 96
18:31:26.533581 arp who-has wardrobe (Broadcast) tell susan
18:31:26.534819 arp who-has wardrobe (Broadcast) tell susan
18:31:28.364162 arp who-has susan (Broadcast) tell susan
18:31:30.394055 arp who-has susan (Broadcast) tell susan
18:31:32.384176 arp who-has susan (Broadcast) tell susan
18:31:34.384204 arp who-has susan (Broadcast) tell susan

Also, here is the tail of /var/log/messages:
Jun 16 18:49:38 wardrobe rpc.mountd: authenticated mount request from
susan:514 for /var/solaris9_install/Solaris_9/Tools/Boot
(/var/solaris9_install)
Jun 16 18:49:38 wardrobe rpc.mountd: authenticated mount request from
susan:514 for /var/solaris9_install/Solaris_9/Tools/Boot
(/var/solaris9_install)
Jun 16 18:49:52 wardrobe rpc.mountd: authenticated mount request from
susan:1023 for /var/solaris9_install/Solaris_9/Tools/Boot
(/var/solaris9_install)

As for the NFS linux vs. solaris thing, I tried exporting the file
system as read-write instead of read-only, but achieved the same
result. Similarly, I used the NFS no_root_squash option, but failed
again.

Thanks,
Chris

Quote:> Linux and Solaris NFS are apparantly different in some aspects.  Take a
> look at this thread...

> http://groups.google.com/groups?th=7df5a88654120b61

> msgs 11 and 18-20 may be relevant.

Thanks
Chris

Why don't you use one of the UltraSPARC machines as the install server?

You may be able to turn on debugging on the server side to see if you're
getting any messages about these mounts.

--

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. >