Installing A Cluster

Network configuration

Head nodes works as NAT for slave nodes

eth0 connects to outside world eth1 is internal

Internal network is 10.0.200.0/255.0.0.0

OS installation

Normal server Linux install with the following packages:

• ssh server
• tftpd-hpa
• dhcp3-server
• nfs-kernel-server
• debootstrap

• libpmi
• mpich2
• slurm-llnl

Grab pxelinux from the web

Setting up a chroot for the node root

The chroot is a directory on the head nodes that hosts the root filesystem that the slave nodes will use. A chroot file system is installed with debootstrap and can be chrooted into so it acts like a standard install.

mkdir -p /nodes/lucid
debootstrap lucid /nodes/lucid/ http://backup.iac.isu.edu:9999/ubuntu/
cp /etc/apt/sources.list /nodes/lucid/etc/apt/
cp /etc/resolv.conf /nodes/lucid/resolv.conf
chroot /nodes/lucid/
apt-get update
locale-gen en_US.UTF-8
dpkg-reconfigure tzdata

More information about chroots, especially helpful locale setup.

In the chroot we want to install the following packages:

• slurm-llnl
• munge
• nfs-common
• openssh-client
• openssh-server

NFS

On the head node, add the following to /etc/exports

/nodes/lucid    10.0.200.0/24(ro,async,no_root_squash,no_subtree_check)
/home           10.0.200.0/24(rw,async,root_squash,no_subtree_check)

see man 5 exports for more information. There are also more nfs settings in

/etc/default/nfs-common 
/etc/default/nfs-kernel-server 

Netbooting

Setting up dhcp

Edit /etc/dhcp3/dhcpd.conf as follows:

 # The ddns-updates-style parameter controls whether or not the server will
 # attempt to do a DNS update when a lease is confirmed. We default to the
 # behavior of the version 2 packages ('none', since DHCP v2 didn't
 # have support for DDNS.)
 ddns-update-style none;

 # option definitions common to all supported networks...
 option domain-name "iac.isu.edu";
 #note the comma between dns servers
 option domain-name-servers 134.50.254.5, 134.50.57.57;

 #these short times are for testing only
 default-lease-time 60;
 max-lease-time 120;

 # If this DHCP server is the official DHCP server for the local
 # network, the authoritative directive should be uncommented.
 authoritative;

 # This is a very basic subnet declaration.
 subnet 10.0.200.0 netmask 255.255.0.0  
 {
  range 10.0.34.1 10.0.34.200; #we really don't even need a range
  option routers 10.0.200.1;
  filename "pxelinux.0";
 }
 #eth0 on brems2
 host brems2_0
 {
  hardware ethernet 00:50:45:5C:10:54;
  fixed-address 10.0.200.2;
  option host-name "brems2";
 }

Edit /etc/default/dhcp3-server

INTERFACES=eth1

This will avoid dhcp serving on the outside network!

service dhcp3-server start

Setting up tftp

Edit /etc/default/tftpd-hpa

RUN_DAEMON="yes" #had problems with inetd in the past
OPTIONS="-l -a 10.0.200.1 -s /var/lib/tftpboot"

Scheduler installation

The Quick Start Administrator Guide is very helpful.

1. Install

• slurm-llnl
• slurm-llnl-slurmdbd
• slurm-llnl-doc
  • mkdir /var/run/slurm-llnl

Munge

Munge is an authentication framework recommended by slurm. All the configuration it needs is:

root@brems:# /usr/sbin/create-munge-key
Generating a pseudo-random key using /dev/urandom completed.
root@brems:# /etc/init.d/munge start