Overview of the setup process

Db2® Mirror is initially configured on a single partition called the setup source node. During the setup and configuration process, the setup source node is cloned to create the second node of the Db2 Mirror pair called the setup copy node. The setup copy node is configured and initialized automatically by Db2 Mirror during its first IPL.

Start of change The Db2 Mirror configuration and setup process supports both external and internal storage. For external storage, the copy services functions available on external storage systems are used during the cloning process. IBM® or non-IBM external storage systems can be used, but IBM storage systems are recommended because the cloning process is automated. Non-IBM storage systems require manual steps to be taken on the storage system to perform the cloning process. Setting up Db2 Mirror on systems using internal storage is also a manual process. End of change

Db2 Mirror automates the cloning for the following IBM storage systems:

IBM Storage Virtualize family
IBM System Storage® DS8000® family

The cloning technologies used for IBM storage systems are FlashCopy® and remote copy. FlashCopy is used when both Db2 Mirror nodes connect to the same IBM storage system. Remote copy is used when the Db2 Mirror nodes are connected to different IBM storage systems.

Start of change The process for setting up Db2 Mirror on systems with internal storage is a manual process. The system objects and all licensed programs on the source node must be manually saved to a tape or optical device. The copy node is then manually booted and installed from the save media. Db2 Mirror automatically configures and initializes the copy node and then clones all replicated objects by adding them to the object tracking list and begins synchronization of the objects. End of change

Both cold cloning and warm cloning are supported by the Db2 Mirror setup process. A cold clone requires the setup source node to be shut down during the cloning portion of the setup process. A warm clone allows the setup source node to remain active during the entire Db2 Mirror setup and configuration process.

Db2 Mirror synchronously replicates data between two IBM i nodes. The Db2 Mirror environment consists of those two nodes plus additional nodes for managing the environment depending on how you choose to setup and manage the Db2 Mirror environment. The recommended environment consists of three IBM i nodes, as seen in the figure below.

During the setup and configuration of Db2 Mirror, the three nodes are referred to with the following names:

Managing Node

The managing node, sometimes called the third node, is where you initiate and manage the Db2 Mirror setup process by either launching the setup wizard in the Db2 Mirror GUI or calling the db2mtool command line tools.

You must use a third IBM i node as the managing node while setting up Db2 Mirror.

It is recommended that you continue using the third node after setting up Db2 Mirror to manage the environment. Since the Db2 Mirror GUI is a web based interface, it requires http or https to be accessed. Using a third node keeps this http traffic off your production servers. Another reason to use a third node is to protect the reliability of the quorum data by adding the third node to the same cluster as the Db2 Mirror nodes. See Clustering for more details.

Setup Source Node

The setup source node is an IBM i partition that is often an already established production partition where your applications are running. The setup process will make a copy of this node and establish the Db2 Mirror environment. After the setup process is complete, the setup source node will be the primary node.

Setup Copy Node

The setup copy node is created by the setup process. It is created by cloning the setup source node and then configured by Db2 Mirror automatically during its first IPL. After the setup process is complete, the setup copy node will be the secondary node.

Db2 Mirror setup process steps

The complete Db2 Mirror setup process can be described by the following five high level steps:

Step 1: Pre-configuration

Before starting to configure Db2 Mirror, you must do the following:

Read and understand the planning considerations. See Planning considerations.
Verify all environment requirements. See Planning considerations.
Gather required information needed during the setup process. See Planning checklist.
Create the setup copy partition. See Hardware Management Console (HMC) requirements.
Configure the storage. See Storage requirements.

Step 2: Configure Db2 Mirror settings

The easiest way to configure Db2 Mirror is by using the setup wizard that is part of the Db2 Mirror GUI. This will guide you through all the steps and control the cloning process used to create the Db2 Mirror pair. See Setting up Db2 Mirror using the GUI setup wizard.

Alternatively, you can manually set up Db2 Mirror using SQL services and other command line interfaces provided by Db2 Mirror. See Setting up Db2 Mirror using commands.

Step 3: Verify Db2 Mirror configuration is complete and valid

After the configuration choices are made, but before the cloning process can be performed, a validation step must be done to verify that Db2 Mirror has been successfully configured. This process performs the necessary checks to ensure the clone process will be successful and that active replication can begin successfully after the clone.

The validation step is done automatically at the end of the Db2 Mirror setup wizard, described in Validation and cloning.

If not using the GUI, the validation step is completed by using the db2mtool command with the precheck action, described in the db2mtool command reference action=precheck.

Step 4: Perform the cloning process

After the validation step is completed successfully, the cloning process can begin. The setup source node is cloned to create the setup copy node. The setup copy node is configured and initialized automatically by Db2 Mirror during its first IPL.

The cloning process will begin automatically at the end of the Db2 Mirror GUI setup wizard, described in Validation and cloning.

If not using the GUI, the cloning process can be started by using the db2mtool command with the start action, described in the db2mtool command reference action=start.

Step 5: Verify clone completed successfully

The final step after cloning is another validation step. This validation step verifies the cloning process was successful and active replication between the two Db2 Mirror nodes began successfully.

The validation step will begin automatically at the end of the Db2 Mirror GUI setup wizard, described in Validation and cloning.

If not using the GUI, the validation step can be performed using the db2mtool command with the checksysbas action, described in the db2mtool command reference action=checksysbas.