[Oscar-users] Nodes hang during network book

[Edmund Bertschinger]

Hi,

I have a problem during systemimager installation of software in OSCAR 2.1 during the stage of network booting and loading the nodes:

I have a 7-node cluster consisting of nodes with 1.7GHz P4 Socket 478 processor and Intel D850MVL motherboard which includes the Intel Ether Express Pro 10/.100+ network card.

The short summary of the problem is that rsync on any client (i.e. any of the 7 compute nodes) cannot find the network to copy files from the server, despite the fact that the client pings the server without trouble.
The server's IP is 10.0.0.255 while this particular node has IP 10.0.0.245

Here are the last 39 lines lines that appear on the node terminal, after a successful PXE boot:

DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 5
DHCPDISCOVER on lo to 255.255.255.255 port 67 interval 67
DHCPOFFER from 10.0.0.255
cat: /floppy/local.cfg: No such file or directory
cat: /floppy/local.cfg: No such file or directory
DHCPREQUEST on eth0 to 255.255.255.255 port 67
DHCPACK from 10.0.0.255
cat: /floppy/local.cfg: No such file or directory
SIOCADDRT: Network is unreachable
bound to 10.0.0.245 -- renewal in 21600 seconds.

Pinging image server "10.0.0.255"
to ensure we have network connectivity

Ping attempt number 1:
PING 10.0.0.255 (10.0.0.255): 56 data bytes
64 bytes from 10.0.0.245: ICMP_SEQ=0 ttl=255 time=0.0ms

--- 10.0.0.255 ping statistics ---
1 packets transmitted, 1 packets received, 0% packet loss
round-trip min/avg/max = 0.0/0.0/0.0 ms

We have connectivity to the Image Server!

Using rsync to copy 10.0.0.255::scripts/host...
failed to connect to 10.0.0.255 -- Network is unreachable

Trying to get hostname via DNS...

Nameserver not running
10.0.0.245 PTR record not found, try again

Couldn't find this hosts name in /tmp/hosts or via DNS!!!

<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Your autoinstall has failed. Please send the above messages to
[EMAIL PROTECTED]

Please press Enter to activate this console
[end of messages]

At this point, I hit Enter to go into console mode to try to see what was going on. A kernel had been loaded into ramdisk. There was no ssh or rsh anywhere; could this explain why rsync couldn't connect? On the other hand, there was telnet, and when I tried to telnet to the server, I got the same response: Network is unreachable. Yet pinging the server works fine! And I'm running xinetd on the server with all the relevant services.

How can I get the client to connect with the server? Can anyone help me with this?

Thanks,
Ed



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Oscar-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-users