Booting Diskful Compute Nodes¶

In addition to booting diskless clients, ClusterWare can also integrate with "diskful" compute nodes that boot from full installations on local disk drives. See Using Kickstart and Appendix: Using Debian Preseed for examples.

Add a locally installed node to the cluster using the same mechanisms as a diskless node. For example, if the following are true:

The new node's network interface is physically connected to the private cluster network shared by existing head node(s) and compute nodes
The interface is configured to use a dynamic IP address assigned by DHCP at boot time

then execute the following command on the head node:

scyld-nodectl create mac=00:11:22:33:44:55

ClusterWare assigns the next available IP address to that MAC address.

Alternatively, if the new node has a static IP address, first ensure that static address is in the defined ClusterWare DHCP range, then specify that static address in your command:

scyld-nodectl create mac=00:11:22:33:44:55 index=100 ip=10.10.42.100

Installing the clusterware-node Package¶

For the new node to fully integrate as a ClusterWare compute node, install the clusterware-node package and dependencies on the new node as the root user. The node does not use the initramfs provided by a ClusterWare boot configuration for diskless nodes. As a result, after installing clusterware-node, update the configuration file /opt/scyld/clusterware-node/etc/node.sh on the node. The configuration file instructs the node how to communicate with a head node. This file must define the base_url of a head node and optionally the iface network interface name that assigned the MAC address used in the scyld-nodectl create command that added the node to the database.

For example, for interface eth0 that connects to head node head-01, make the following edits:

[root@newnode]$ cat /opt/scyld/clusterware-node/etc/node.sh
# Specify a base_url for any head node
base_url=http://head-01/api/v1

# Specify the network interface used to reach the head node(s)
iface=eth0

# Optionally force (sslverify=yes) or disable (sslverify=no) SSL
# certificate checking. Leaving this option blank allows for HTTPS
# without certificate checking.
#sslverify=

On the node, run the following command:

/usr/bin/update-node-status --hardware --upload

This command sends the initial node hardware information to the head node. You can then reboot the node to fully integrate it into the cluster.

The newly booted compute node is controlled through the customary scyld-nodectl reboot, shutdown, and exec commands. To support --hard mode, configure the node's power_uri field to provide appropriate ipmitool authentication and an IP address of the node's Baseboard Management Controller (BMC). For example: ipmi:///admin:password@172.45.88.1. See Compute Nodes IPMI Access and Database Objects Fields and Attributes for details.

Additional Support for Diskful Nodes¶

ClusterWare head nodes provide additional support for diskful node installation through a set of pre-installer scripts, installer scripts, and a repository of packages that compute nodes can use to download and install node packages. Additionally, the head node repository can automatically convert some packages to a TAR file that should be installable on a wide variety of platforms.

See Support for Diskful Nodes for details, including the preparation steps needed on the head nodes.