Isolating Network, Node, And Linux Problems; Troubleshooting The Shared Vlan - IBM Cluster 1350 Installation And Service

Hide thumbs Also See for Cluster 1350:

Table of Contents

v The customer may elect to configure a subset of Cluster Nodes with additional

v A supported distribution of the Linux operating system.

v Cluster management software such as CSM.

CSM maintains a database of configuration information about the nodes that are

configured in the Cluster 1350. The following CSM command run on the

management server will display this node information: lsnode -Al

The output of this command provides a great deal of information about each node

such as its type, model number, serial number, and hostname. The output also

provides important information that relates each node to the terminal server

network and service processor network. For the terminal server network, the

output includes the console server hostname and the console port number to

which the node is connected. For the service processor network, the output

includes the hostname of the Falcon card to which the node is connected and the

internal service processor name for the node.

CSM distinguishes between the management node, pre-managed nodes, and

managed nodes. Pre-managed nodes are nodes that have been added to the

configuration but are not yet ready to be managed, for example, because they have

not yet been installed. Running lsnode -P on the management server will generate

a list of pre-managed nodes in the configuration while lsnode only lists the

managed nodes. Running mgmtsvr from any node will display the management

server.

Isolating network, node, and Linux problems

Cluster 1350 nodes are connected over a 10/100 Mbit Ethernet Cluster network. A

Cluster 1350 may also have a second network, either an additional Ethernet

network or a Myrinet 2000 network.

As a preliminary diagnostic step, ping all the nodes over all available networks.

Note the results and compare the symptoms seen to Table 11

Installation and Service

external storage - for example as Storage Nodes for GPFS. This may be a Fiber

Channel solution (using a FAStT storage subsystem).

Table 11. Troubleshooting the shared VLAN

Symptom

Action

Table of Contents