Home > Backup and Disaster Recovery Resources from Dell AppAssure > Dell AppAssure: Backup and Recovery Tech Docs > Replay 4 – Server Recovery Point Replication

Replay 4 – Server Recovery Point Replication

TECHNICAL OVERVIEW

Introduction

This document describes the Replication feature of Replay 4. Replication is designed to replicate
recovery points between Replay Cores in an efficient and safe manner to enable off-site backup and offsite
disaster recovery.

Replay Replication option can be enabled or disabled on a per protected server basis. This option is
available in single server and multi-server implementations. The cool thing about Replay Replication is
that it replicates the compressed and deduplicated recovery points over the WAN so its bandwidth
efficient saves storage on the LAN and at the DR location.


Figure 1 – Off-Site Disaster Recovery

How it works

This section describes how Replay Replication works by first defining our terminology followed by a
review of our approach.

Terms

  • Protected Server – A Windows workload, physical or virtual, that is protected by a Replay Agent.
    Supported environments include Windows® Server 2008 R2, Windows® 2008 Server, Windows®
    2003 Server, Windows 7, Windows Vista, Windows XP.
  • Replay Core– processes and stores compressed and deduplicated recovery points as
    incremental forever images of the protected servers. Cores can be installed on the protected server or on a dedicated server or worker virtual machine. This role performs many functions
    including creating virtual standby environments, application validation, restores and replication.
  • Recovery Points – Represents a point-in-time image of the protected server. They are stored as
    compressed and deduplicated files on storage available to the Replay Core, DAS, NAS, SAN.
  • Replication Source – The server from where the replication of the recovery points is initiated.
    Replication is always performed between two Replay Cores.
  • Replication Target – The server that is the destination of the replicated recovery points.

Snapshots

Replay 4 creates point-in-time snapshots of the protected servers and they are stored as compressed
and deduplicated recovery points in a directory accessible from the Replay Core. The recovery points
are maintained on a per protected server basis in a directory structure as follows:

Drive:\TevoRepository\ProtectedServerName1
recoverypoint4Incremental
recoverypoint3Incremental
recoverypoint2Incremental
recoverypoint1Base

\\ShareName\TevoRepository\ProtectedServerName2
recoverypoint4Incremental
recoverypoint3Incremental
recoverypoint2Incremental
recoverypoint1Base

Replication

Replication copies the recovery points from a replication source to replication target on a per protected
server basis. As recovery points are replicated, all recovery point files and registry settings are replicated
to the secondary Replay Core, such that the loss of the replication source won’t mean the loss of all
recovery points and protection settings. In the case of a Replay Core outage, the Replay Core can be
rebuilt, and resynchronized with a mirror, though until resync is complete the primary Replay Core
won’t be able to accept new snapshots.

Here are some facts about Replay Replication:

  • As new snapshots are taken on the replication source, they will be replicated to replication
    target.
  • Replication is performed on a per protected server basis.
  • Recovery points are replicated oldest-first, to ensure the replication target always has a valid
    and mountable epoch chain.
  • Recovery point changes due to rollup will not be replicated reducing the amount of data that
    needs to be transferred or the WAN. The replication target will perform its own rollup on the
    replicated recovery points.
  • Rollup and replication will not conflict with one another.
  • It is assumed that the replication target is on a different subnet and behind a different firewall
    than replication source, thus the replication target will only communicate with the agents
    protected by replication source for rollback purposes.
  • The replication target is able to perform all normal Replay restore functions such as BMR,
    instant rollback, VM export, even if the replication source is down.
  • If Exchange is being protected, Exchange and system DLLs from the replication source will be
    propagated to the replication target so features such as force mountability check, MR, and P2V
    are available on the replication target.
  • Mismatched versions between replication source and replication target are detected.
  • The replication source and replication target can be any combination of supported operating
    systems.
  • In the event of a link failure or transfer error, when the link comes back up, the transfer is
    resumed from where it was interrupted.
  • All recovery points that are replicated are verified for integrity.

Replication Implementation Details

  • Internally, the Replay Cores communicate over HTTP. The target listens on port 8080 for HTTP
    connections from the source. All communications are initiated by the source to the target.
  • Replay uses the adler-32 checksum algorithm to verify the integrity of transferred files, and
    detect when a partially-transferred file can be resumed without data loss.
  • To determine what files to replicate, Replay chooses all source files not yet on the target, and
    those files on both which have not yet been rolled up but have different metadata between the
    two sides.
  • Each file is transferred in a separate HTTP PUT operation. Once an entire epoch is transferred, it
    is loaded into Replay and available for use. Thus, some replicated recovery points are available
    for use even before all recovery points have replicated.

Replication Topologies

Replay support various replication topologies to meet your business needs.

1 : 1 Configuration

The 1 : 1 configuration is that standard configuration and is useful for protection of a single server or
group of servers from 1 site to another.


Figure 2

Many : 1 Configuration


Figure 3

Many: Many Configuration

Replay Replication supports replication on a per protected server basis. This means that you can
replicate different protected server’s recovery points to different replication targets in different
locations as shown in the figure below.


Figure 4

Replication Schedules

Replication can be scheduled to run during certain times during the day. Any time outside of this
window, replication will be suspended. The schedule is specified as a start and stop time for weekdays,
and a start and stop time for weekends.

  • Users can, on an ad-hoc basis, pause a currently running replication for an hour, a day, or until resumed
  • Users can configure VM export, recovery point export, or continuous rollback on the replication target for recovery points replicated from the replication source.

Recovery Point Copy and Consume

The copying feature is useful when the WAN link is insufficient to move large amounts of data. It copies
the existing recovery points including the base image from the replication source to a local removable
device. Once the copy is complete, the removable drive can be inserted in the replication target and
consumed. The replication target is seeded with the recovery points from the replication source
including the base image. The replication will resume by replicating the new recovery points only.

Authentication between Replay Cores

Replay uses NTLM authentication over HTTP. Once authenticated between the replication source and
the replication target, Replay generates a one-time-use key, which it exchanges via RSA over HTTP. Each
request sent to the replication target over HTTP is authenticated with a separate one-time key. Thus authentication is hardened against Replay attacks, but potentially vulnerable to a man-in-the-middle
attack since we have no way of authenticating the identity of the remote server.

Back To Top

CURRENTLY READING

Replay 4 – Server Recovery Point Replication

Download PDF

Back to the List of Tech Docs

In This Section

See Also


Installation & Evaluation Guides

Deployment Guides

User Guides

MailRetriever

DocRetriever



Dell AppAssure Software for Exchange continuously protects and monitors the health of your Exchange data stores and allows administrators to quickly search, recover, and analyze mailbox content. With Dell AppAssure Software for Exchange you can backup and restore individual email messages, folders, or mailboxes to a live Exchange server or directly to a PST, thereby solving some of your most costly and time consuming challenges.
Dell AppAssure Software for SQL is an enterprise SQL server backup and disaster recovery software solution developed to assure comprehensive protection - including system and user databases, SQL binaries and the server - making it possible to restore both servers and databases in just minutes.
Dell AppAssure Software for Hyper-V is an enterprise-class backup and disaster recovery software solution designed for Windows Server 2008 Hyper-V environments. Dell AppAssure Software for Hyper-V supports both host and guest level backup methodologies while delivering accelerated backups, fast local recoveries and disaster recovery, all-in-one solution.
Dell AppAssure Software for SharePoint is an enterprise backup and disaster recovery software solution developed to assure comprehensive protection - including front-end servers, SQL servers and the web servers - making it possible to restore both the entire SharePoint environment, individual SharePoint servers and databases and individual SharePoint objects in just minutes.
Dell AppAssure Software for Windows Servers is an application-aware, server-based backup and disaster recovery software solution that delivers fast application backups, fast local recoveries and off-site disaster recovery, all-in-one solution.
Dell Small businesses typically have limited IT resources and smaller staffs than large enterprises, but with Dell AppAssure Software for Windows Small Business Server they have full access to the same powerful server backup and recovery software tools found in larger organizations.
Dell AppAssure Software for VMware is an enterprise-class backup and disaster recovery software solution designed for Windows workloads running VMware virtual environments. With one price for an unlimited number of guests, Dell AppAssure Software delivers accelerated backups, fast local recoveries and disaster recovery, all-in-one solution.
Dell AppAssure Software for Windows Desktops (VDI) simplifies and automates backup and disaster recovery processes, while giving an administrator the ability to remotely monitor and manage large VDI deployments or geographically distributed machines from a single console.
DocRetriever for SharePoint dramatically simplifies the process of restoring individual documents, sites, sub-sites and folders. SharePoint Administrators no longer need to spend countless hours rebuilding your entire production database, or worry about deploying a temporary SharePoint recovery farm, just to recover an individual document or list item.
MailRetriever for Exchange is an affordable and easy-to-use e-mail recovery software package for browsing, recovering, searching, exporting, and analyzing e-mail and mailbox content directly from any unmounted Microsoft Exchange Data Store (EDB). MailRetriever for Exchange can even open and repair corrupted EDB files. With MailRetriever for Exchange, you are never more than just a few clicks away from recovering a message or mailbox. MailRetriever for Exchange eliminates the need for time-consuming full-server and EDB restores for message recovery.
For individual message or mailbox recovery, Dell AppAssure Software and Microsoft Storage Solutions have partnered together to deliver MailRetriever for DPM, enabling Microsoft Exchange administrators around the world to gain efficient and economical access to message-level recovery.
Dell AppAssure´s Backup software for Kaseya lets you recover from any server failure to physical or virtual machines in minutes, backup and recover either locally or offsite, and provide offsite DR to customers with low bandwidth through integrated deduplication and replication.
Dell AppAssure Backup & Disaster Recovery Software is best way to back up and protect your VMware virtual machines. It doesn´t just support your virtual environment - rather; it leverages VMware ESX, ESXi and vSphere to provide a whole new level of protection for all your virtualized applications, services and desktops. Dell AppAssure Software lets you meet your VMware backup RTO and RPO objectives for less.