IBM CSM to IBM Systems Director Transformation Guide

44 IBM CSM to IBM Systems Director Transformation Guide

Figure 3-2 CSM environment for the coexistence scenario

3.2.2 Cluster verification

Before starting the transformation, it is important to perform functional verification to ensure

your cluster is working correctly.

򐂰 List the nodes that are managed by CSM, as shown in Example 3-1.

Example 3-1 List the CSM-managed nodes

p5570lp01(root)/> lsnode

p5570lp02

p5570lp03

p5570lp04

p6lp01

p5570lp01(root)/>

Managed System

Managed System

Power 5

Power 6

HMC1 - CSM

Hardware Control

Point

CSM Cluster

HMC2 - CSM

Hardware Device

Virtual Server: p61p01

OS: AIX 6.1 TL 06 SP 05

Role: CSM client

Virtual Server: p55701p04

OS: AIX 6.1 TL6 SP5

Role: CSM note

Virtual Server: p55701p03

OS: AIX 6.1 TL 06 SP 05

Role: CSM note

Virtual Server: p55701p02

OS: AIX 5.3 TL 12 SP 01

Role: CSM note

Virtual Server: p55701p01

OS: AIX 6.1 TL 06 SP 01

Role: CSM & NIM Srv

Virtual Server: p5570vio1

VIOS 2.2.0.11-FP-24 SP-01

Role: CSM HW Device

CSM Logical Network

HMC Logical Network

Note: CSM executable commands can be found in the /opt/csm/bin directory. That

directory is in the PATH in the following examples, as typically recommended during CSM

implementation. Therefore, full path names are not used.

Chapter 3. Transformation scenarios 45

The 3 nodes p5570lp02, p5570lp03, p5570lp04 are the POWER5 LPARs and node p6lp01

is the POWER6 LPAR.

򐂰 List the hardware devices that are managed by CSM as shown in Example 3-2.

Example 3-2 List all the CSM-managed hardware devices

p5570lp01(root)/> lshwdev

hmc2

p5570vio1

p5570lp01(root)/>

The HMC that manages the POWER6 server is hmc2 and p5570vio1 is the VIO Server for

the POWER5 server.

򐂰 The node group CSM1_nodegroup contains all the CSM-managed nodes. See

Example 3-3.

Example 3-3 Show the members of the CSM node group

p5570lp01(root)/> nodegrp CSM1_nodegroup

p5570lp02

p5570lp03

p5570lp04

p6lp01

p5570lp01(root)/>

򐂰 The csmstat command displays the status of each node, as shown in Example 3-4.

Example 3-4 Show the status of the CSM-managed nodes

p5570lp01(root)/> csmstat

---------------------------------------------------------------------------------

Hostname HWControlPoint Status PowerStatus Network-Interfaces

---------------------------------------------------------------------------------

p5570lp02 hmc1 on on en0-Online en1-Online

p5570lp03 hmc1 on on en0-Online en1-Online

p5570lp04 hmc1 on on en0-Online

p6lp01 hmc2 on on en0-Online

p5570lp01(root)/>

In many CSM environments, customer utilize the distributed condition-response

capabilities. We have created a condition and a response to demonstrate this.

򐂰 The condition Filesystem_Tmp_Monitor1 is defined for the node p5570lp02. This condition

sends an email to the root user when the /tmp file system utilization exceeds 90%. The

condition gets rearmed when the percentage utilization drops below 75%. Refer to

Example 3-5.

Example 3-5 List the details of the distributed RMC condition

p5570lp01(root)/> lscondition Filesystem_Tmp_Monitor1

Displaying condition information:

condition 1:

Name = "Filesystem_Tmp_Monitor1"

Node = "p5570lp01"

MonitorStatus = "Monitored"

ResourceClass = "IBM.FileSystem"

EventExpression = "PercentTotUsed>90"