Exploring Storage Management Efficiencies and Provisioning - Understanding IBM TotalStorage Productivity Center and IBM TotalStorage Productivity Center with Advanced Provisioning

174 Exploring Storage Management Efficiencies and Provisioning

7.6 Executing existing workflows

There are two options that you can use to create a new workflow. You can use

the Web-based user interface, or you can write your workflow in any text-based

editor and import the workflows and Java plug-ins into IBM Tivoli Provisioning

Manager. In this example, we will use the Web-based interface.

You can use the Logical Device Operations to automate the lower level tasks to

provision storage. This will keep the Data Center Model (DCM) up to date, and it

also abstracts multiple steps into a single call. Some of the more common work

flow scenarios we have seen in customer environments are:

򐂰 Adding storage to a server (assuming that the zones do not need to change)

򐂰 Adding storage to a server and manipulating the zones

򐂰 Adding storage to storage pools and assigning it later to a server

And you can also do the opposite — remove capability for each one of those

activities, or reverse what has been created to free up storage.

We will start by describing high-level activities needed to make all this possible.

Before beginning, we need to verify that IBM TotalStorage Productivity Center

with Advanced Provisioning is properly installed, which includes the enablement

layer. Install IBM Tivoli Provisioning Manager and import the workflows and Java

Plug-ins.

After installation, the first thing to do is to execute the TPC_Configure and

TPC_Initialize workflows. Today, IBM Tivoli Provisioning Manager by itself does

not have automatic device discovery capability, so this initialization workflow

imports information on storage systems, storage pools, and fabric information as

well as host HBA information, into IBM Tivoli Provisioning Manager. It does this

by activating workflows that go out and query both IBM TotalStorage Multiple

Device Manager and IBM Tivoli SAN Manager, bringing back the needed

information and storing it in the DCM. Once you do this, you can look at SAN

environments, subsystems it has discovered, servers, HBAs these servers have,

etc., in IBM Tivoli Provisioning Manager. All of these devices will be represented

as objects in the IBM Tivoli Provisioning Manager Web-based interface.

Before we can go into the details of how to use the tool to create custom

workflows, we need to discuss how to address the components that have been

discovered. IBM Tivoli Provisioning Manager gives an Object Identification

Number (ObjectID) to each device or object — whether logical or physical —

within its configuration database. First, navigate to the Data Center Assets and

Resources tab, an icon that is the first tab at the top in the left pane of the IBM

Tivoli Provisioning Manager display. Refer to Figure 7-11 on page 180 — this is

the tab that looks like a stack of three boxes.

Chapter 7. Workflows 175

These ObjectIDs can be seen in the product by placing the mouse pointer over

the item in the left pane of the display. See the number ‘1125’ next to the mouse

pointer in Figure 7-5; this means that the ObjectID for the ‘COLORADO_HBA:0’

is ‘1125’. This number will be used to represent the object in the input screens.

Figure 7-5 Display of the ObjectID

In the following example workflow, we want to accomplish the tasks listed in that

same order. We would also like to show you the underlying workflow language

that drives the workflow. The following tasks will be demonstrated:

1. Gather needed device data

2. Create storage needs

3. Create storage volume

4. Map storage volume to server (or “fabric adapter”)

5. Create the zone

Gather device data: Write down the IBM Tivoli Provisioning Manager object IDs

for all the devices/components/servers that will be involved. The object IDs are

how each object (device) is recognized within IBM Tivoli Provisioning Manager.

This includes the servers, storage devices, storage pools, and the fabric switch.

Important: Be aware that the ObjectID numbers will change if you rerun the

initialization workflow.