IBM ProtecTier Implementation and Best Practices Guide

Chapter 21. Replication and disaster recovery overview 335

ProtecTIER V3.2 and later (machine 3958-DD5) supports native replication in all available

functional modes:

򐂰 Virtual Tape Library (VTL)

򐂰 OpenStorage (OST)

򐂰 File System Interface (FSI)

21.2 Disaster recovery operations

Disaster recovery is the process of recovering production site data in a DR location. Disaster

recovery is useful if a disaster occurs or a situation occurs where the production (or primary)

site goes offline. The hub, or DR site, can take the place of the production site until the

primary site comes back online. When the primary site comes back online, previously

replicated and newly created tapes can be moved to the main production site using the

failback process. The primary site then resumes its roll as production and primary site for

the user.

When the user enters the ProtecTIER Disaster Recovery (DR) mode, all incoming replication

and visibility switching activities from the failed production site to the DR site (hub) are

blocked. When that primary site is rebuilt or replaced, the user can then return the data to that

primary site and continue the backup and replication operations. If necessary, the user can

use the DR repository as a backup target. The user must enter the DR mode at the DR

repository to start working with the DR site as their primary site.

21.2.1 Entering ProtecTIER DR mode

Keep the process for entering DR mode simple to avoid potential complications and human

errors when handling an emergency situation. The special DR wizard in ProtecTIER Manager

at the DR site facilitates the initiation of DR mode. DR mode blocks all replication activities

(including visibility switch) at the DR site from the failed primary site (and at that primary site if

it is still up).

Network failure: If a network failure occurs during replication, the system tries, up to

seven consecutive days, to replicate any specific cartridge that did not finish its replication

because of the line failure.

336 IBM ProtecTIER Implementation and Best Practices Guide

Figure 21-1 shows an example of entering ProtecTIER DR mode.

Figure 21-1 Entering ProtecTIER DR mode

As shown in Figure 21-1, you complete the following steps to entering the mode:

1. From the Systems Management view, click the destination repository.

2. Click Replication  Replication Disaster Recovery  Enter DR Mode.

3. Choose the repository you want to enter DR mode and confirm the selection.

An automatic procedure is run that blocks incoming replication to the DR site from the

source selected.

21.2.2 Working at the disaster recovery site

Replicated cartridges at the DR (hub) repository that are created at a primary site (spoke) are

in read-only mode. If it is necessary to run local backups at the DR site, and use new, locally

created cartridges.

Failback

Failback is the procedure for replicating back updated cartridges, new or old, from the DR site

to the original (or restored) production site to bring it up-to-date in case it was down, or lost

and rebuilt. You use failback to return to the original working mode of the primary site.

Important: Entering and activating ProtecTIER DR mode allows you to have all tape

cartridges with the latest data visible at the remote site with read/write access. However,

the backup application must be restored manually or by using prepared semi-automatic

procedures and scripts. You must be familiar with this backup application

recovery process.

Failback: The failback procedure can be initiated only in DR mode. It is mandatory to

perform the failback process before leaving DR mode at the remote disaster site.