Though the technological reasons to protect Exchange systems may be self evident, there are quantifiable fiscal reasons to protect these messaging systems as well. The dollar value of any given data-set may be difficult to calculate, but the cost-savings of avoiding even a single Exchange outage can easily be determined. In most organizations, there are at the very least one or two subsets of end-users who cannot continue to work without these systems functioning.
Even if these groups might not regularly produce revenue in the form of direct sales or billable engagements, salaries, benefits and fixed costs still accrue during an outage. Therefore, the loss of the messaging systems for even a few hours could easily result in thousands of dollars in budget outlay without recouping a single dollar in productivity. For revenue-generating groups, the cost of this downtime is even more easily quantifiable. Avoiding even one of these outages is not only a good idea for the IT department, but for the CFO's office as well.
www.doubletake.com888-674-9495PROTECTING MICROSOFT EXCHANGEExecutive SummaryE-mail, shared calendars and contacts are used to power some of your most important business-critical applications. In fact, these systems have gone from being a "nice to have" to being businesstools that are relied on by the entire company from the warehouse to the executive boardroom.Regardless of your job responsibilities, you most likely have an interest and dependence on e-mail.Without it, productivity literally comes to a crawl and can even stop completely.While it was oncejust a convenient way for employees to communicate internally, today e-mail systems like Exchangeare tightly integrated with other important business applications and are one of the primary methodsfor communicating with current and prospective customers. Protecting Exchange systems againstcostly downtime has become a top priority for most IT departments. So then, the question becomes"How do I ensure that my Exchange environment is always protected?"WHITE PAPERBob Roudebush,Director of EngineeringSolutions Published: May 2006Though the technological reasons to protectExchange systems may be self evident, there arequantifiable fiscal reasons to protect thesemessaging systems as well.The dollar value ofany given data-set may be difficult to calculate,but the cost-savings of avoiding even a singleExchange outage can easily be determined. Inmost organizations, there are at the very leastone or two subsets of end-users who cannotcontinue to work without these systemsfunctioning. Even if these groups might notregularly produce revenue in the form of directsales or billable engagements, salaries, benefitsand fixed costs still accrue during an outage.Therefore, the loss of the messaging systems foreven a few hours could easily result inthousands of dollars in budget outlay withoutrecouping a single dollar in productivity. Forrevenue-generating groups, the cost of thisdowntime is even more easily quantifiable.Avoiding even one of these outages is not onlya good idea for the IT department, but for theCFO's office as well.Complete protection of Exchange requires a lotmore than management of on-site and off-site tapebackups.While traditional tape backup is anexcellent tool for long-term archiving, true recoveryfor the sake of business continuity requires real-time data protection which enables disasterrecovery and high availability. Having a solutionthat is cost-effective, hardware independent andscalable is something every IT manager shouldseriously consider.This whitepaper is meant to provide a brieftechnical overview of Microsoft Exchange Server2000 and 2003 survivability requirements as wellas present how a solution such as Double-Take can meet your company's availability and dataprotection needs for Exchange.The need toprotect systems and data is readily apparent.The remainder of this document will cover:qWhat Exchange data needs to be protectedfor business continuityqHow Double-Take protects Exchange dataqHow to ensure Exchange availability- With Stand-Alone Exchange Servers- With Front-End/Back-End Exchange systems- With Microsoft Cluster Services and ExchangeqOther considerations for ExchangeUntitled DocumentWHITE PAPER:PROTECTING MICROSOFT EXCHANGEWhat Exchange data needs to be protected forsurvivability? As of this writing,there are two supported versions of Exchange inuse (2000 and 2003) and those versions of Exchange can run onvarious configurations of Windows 2000 and 2003.For each ofthese Exchange/OS permutations,one must also consider theadditional complexity of service packs and hardware platformsavailable.The result of these permutations of configuration anddeployment options is a wide variety of systems to be protected.The only common denominator in all of these configurations is thatthe various Exchange files are stored on Microsoft Windows filesystems.Each Exchange system is actually a complex configuration ofmultiple databases that are used to store and manipulate dataalong with a message transport system to move e-mail and otherinformation into and out of these databases.Protection of thedatabase files,along with their log files and checkpoint files,iscritical to the recovery of any Exchange server.Generally speaking,each group of databases (Storage Group) will have a set of log filesand a checkpoint file.They are identified by the extensions .LOGand .CHK,respectively.Each database unit (Mail or Public FolderStore) will have an e-mail database (.EDB) and a streaming data file(.STM).Only by protecting all of these files can you be sure you canresurrect an Exchange system that has suffered a disaster.When a recovery from tape is attempted,the tape systems willreplace these files onto the original server or a new server,at whichpoint management tools can be invoked from within the ExchangeSystem and from the command line to re-constitute the databasesback to the state they were in when the tape backup was taken.This is an acceptable way to recover from something like a virusattack or human error where a point-in-time copy is needed torestore a good copy of the data.For a majority of outages likehardware failures and site-wide disasters,however,it is not the mosteffective means of restoring a failed server.In many environments,solutions such as e-mail archiving are alsodeployed as an integral part of a company's e-mail architecture.These tools are designed to remove attachments and outdated e-mails from the Exchange server in order to free up disk space andenhance overall performance.This should be considered as part ofany company's recovery plan for Exchange as it may be possiblethat not all of your vital data actually resides on the Exchangeserver itself.These additional systems will also need to beadequately protected in order to fully restore services to end-users.The same theory applies to Blackberry,GoodLink and other mobileinformation systems which also integrate with Microsoft Exchangeto provide additional functionality such as remote access to e-mail.Without a Disaster Recovery (DR) plan that accommodates thesesystems,restoring the Exchange server itself is only one step on theroad to recovering the entire messaging system.Finally,as with many other database systems,Exchange requiresthat each transaction it writes to disk is performed in an explicitorder.This is tracked continuously so that Exchange can maintainwhich changes have been requested and which have actually beencommitted to the physical database files.Generally referred to as a"transactional database system" and specifically called the JETdatabase engine in Exchange,these databases do not toleratereplication or a backup system that cannot guarantee that write-order integrity of the data is maintained.How Double-Take Protects Exchange DataDouble-Take Software has been protecting messaging systems andother business-critical data since before Exchange 5.5 and evenWindows NT Server 4.0 began shipping.One of the strengths of theDouble-Take replication technology is that it protects files at thebyte-level regardless of the application.In this case,when Exchangewrites data to any of its files,the actual byte-level changes it makesto the Windows file system are sent to another Windows server.Once the data is protected to another server,multiple options areavailable for achieving availability and disaster recovery goals.Theimportant point to consider,however,is that it all starts with thedata - and that means it starts with Double-Take.The first capability this method of data protection offers is a trulyhardware-independent,version-independent,and OS-independentdata protection solution.Simply put,Exchange resides on aWindows file system and Double-Take can protect those file systems.Additionally,by leveraging its patented set of replicationtechnologies,Double-Take can ensure that each Windows I/Otransaction for a protected data-set is not only sent to one or moreDR systems with full data integrity,but also that it will becommitted in the exact same order that the original Exchange servercommitted the changes.Double-Take affords both data and write-order integrity,allowing the Exchange and the JET database systemto immediately recognize a consistent copy of the data on therecovery system(s).2Untitled DocumentWHITE PAPER: PROTECTING MICROSOFT EXCHANGEAn Overview of Double-Take Application Managerfor Exchange.Double-Take Manager for Exchange is designed specifically for theconfiguration and management of Double Take for the protection ofMicrosoft Exchange environments. It automates the setup andconfiguration of replication and failover monitoring of Exchangeservers for rapid and successful deployment. Features such as auto-discovery of servers and key Exchange data files simplify the processdown to just four steps and reduces the risk of human error.Exclusive features such as the pre-flight check help to ensure that allnecessary settings within the environment are configured correctlyfor replication and failover of Exchange. Any errors identified arelisted along with suggestions for resolution. Double-Take Managerfor Exchange checks over 65 different configuration criteria and canautomatically correct a majority of them on the administrator'sbehalf. This helps achieve a seamless, error-proof deployment.Double-Take Manager for Exchange performs recovery and failbackof Exchange servers with a minimal window of downtime. Byperforming recovery tasks while users remain online, Exchangedowntime is reduced to just the few moments that it takes forDouble-Take to stop and start the necessary Exchange services andrelegate processing back to the production server.Double-Take is at the core of our Exchange solutions, makingavailable the extensive set of features and functions it offers. Alladvanced Double-Take features including intelligent datacompression and flexible bandwidth scheduling are available for fastand efficient replication of Exchange data. Double-Take ApplicationManager for Exchange 4.0 further integrates the management ofthe Exchange protection environment into a single console withincreased automation, speed, and efficiency for improved recoverytimes and higher levels of Exchange availability.How to ensure Exchange availability:With Stand-Alone Exchange servers:Exchange systems can come in a variety of flavors andconfigurations to meet nearly every business need and budget. Byfar the most common system employed today is the Stand-AloneExchange server (SAES). SAES systems are a single server runningall components of the Exchange system on the same physical orvirtual system. This means the server acts as a mail-transportsystem, routing system, SMTP server and gateway and mail andpublic folder database server. It may also provide spam filtering,anti-virus scanning or any other function related to Exchange.The end result of this configuration is that you have a single serverthat must be protected - all your eggs are quite literally in onebasket. Double-Take can provide you with data replication of all keyinformation for the Exchange system and any other systems runningon the production server. In addition, the Double-Take ApplicationManager (DTAM) for Exchange can help you prepare a secondaryserver to take over in the event of a loss of the primary. DTAM forExchange provides application-specific configuration, availability andmanagement features and is a free toolkit available as part ofDouble-Take.DTAM for Exchange will select the appropriate directories andvolumes on your production machine which need to be protected,configure Double-Take replication, and configure the secondaryserver's Exchange configuration to match that of the productionmachine. It will also prepare the secondary server to execute thenecessary commands to start Exchange services during an outageand allow failover to occur. During this configuration process, youcan specify what network path the replication systems should use, ifdata should be compressed for transmission and if you wish tomanually initiate the failover process ("one-click failover") or have ithappen automatically after a timeout you define.During an outage, DTAM will either automatically initiate a failoverif the production server is unreachable for the amount of time youset, or alert you and wait for you to initiate the failover manually.Alerting for either type of failover scenario is available via SNMP,SMTP and the Windows Event Log, in addition to the native Double-Take management tools. Regardless of which failover methodologyyou choose, the procedure for restoring services for end-users is thesame.First, Double-Take's DNS Fail Over (DFO) component will updateActive Directory DNS servers to re-route end-users to the recoveryserver. You may specify any and/or all DNS record types for update,depending on what systems you need re-directed. Double-TakeSoftware Professional Services can also assist you with providingautomated failover for customized DNS or Exchange deployments aswell.3Untitled DocumentWHITE PAPER: PROTECTING MICROSOFT EXCHANGEAfter DFO re-routes the end-users, the Double-Take ExchangeFailover (EFO) component will dynamically re-assign all mailboxesand Public Folders from the failed server to the recovery server. Thiswill allow end-users to regain access to their information the nexttime they attempt to connect to Exchange. Internal testing and real-world feedback has revealed that this failover process is very fastand creates a relatively small load on your existing DNS and ActiveDirectory infrastructure - approximately 14,000 users can be movedto secondary server in around 7 minutes. In fact, while conservativefailover estimates are suggested to be placed at 45 minutes for totalfailover, most Double-Take customers report being able to failover inless than 20 minutes - even when performing the failover across aWAN connection.Finally, DTAM will prepare and start the appropriate Exchange and3rd party services (anti-virus, mobile mail, archiving, etc) on therecovery server. This last step re-establishes a live Exchange serverfor your end-users to connect to. Outlook clients may need to bere-started, but no end-user configuration will be required. Inaddition, other components such as Outlook Web Access (OWA)and other e-mail integrated system will simply pick up where theyleft off.How to ensure Exchange availability:With Front-End/Back-End Exchange servers:Front-End/Back-End (FE/BE) configurations for Exchange servers arebecoming more and more popular as spam filters, virus firewalls andother solution sets make putting those systems on a non-databaseserver much more effective. Front-End servers house mail routingand transfer systems, third-party tools for mail management andoften the Outlook Web Access component of Microsoft Exchange.In these configurations, Back-End servers function as repositories formailbox and public folder data used by end users.In many ways, the method by which Double-Take protects FE/BEsystems is very similar to the method employed with SAE systemsdescribed earlier. By leveraging the inherent load balancingcapabilities of Exchange services such as Outlook Web Access or theload balancing feature of other Exchange-related solutions, a secondFront-End server can be configured to accept mail and perform otherfunctions either when the primary Front-End server fails ordynamically as additional resources are needed. Double-Take withthe Double-Take Application Manager for Exchange is then deployedto the Back-End servers to provide disaster recovery and highavailability for the Exchange databases used by the Front-Endservices. All options available for protecting SAE systems withDTAM are also available for protecting Back-End systems.How to ensure Exchange availability:With Microsoft Cluster Services and Exchange Server:In some cases, even a 20 minute failover time is simply too long towithstand. Microsoft Cluster Services (MSCS) can provide yourorganization with faster failover times in some cases - often failoveroccurs in a matter of just a few minutes. Though MSCS providessome benefits in terms of failover for Exchange, there are alsopotential drawbacks to be considered that may affect your decisionto implement the technology. Firstly, MSCS requires hardware thatis certified to be compatible with MSCS and also requires thepurchase of shared disk array. Secondly, the MSCS architecture (twoservers, a shared disk array, etc) may not be ideal for situationswhere you wish to separate the nodes of the customer across greatdistances. Lastly, due to the nature of MSCS and the shared-diskconfiguration it uses, two servers will alternately use the same copyof the data. This makes MSCS an exceptionally good system for localavailability, but potentially introduces a single point of failure if anentire site experienced a disaster or other outage.Double-Take used in combination with MSCS, however, allows forthe protection of an Exchange Cluster to either another single- ormulti-node Exchange cluster at a remote recovery location. Byintegrating seamlessly into the MSCS component of Windows Server,it does this without adding additional complexity to clusteradministration. This is a best-of-both-worlds solution, offeringMSCS failover in the event of a single-server failure, but stillallowing for a redundant copy of the data on another system andthe ability to fail over via DTAM for Exchange to that secondarysystem in the event that the entire production cluster is lost.4Untitled DocumentWHITE PAPER: PROTECTING MICROSOFT EXCHANGEDTAM for Exchange when used in conjunction with an Exchangecluster failover performs nearly identically to the Stand-Alone orFront-End/Back-End configurations outlined earlier. In the case ofExchange clusters, however, replication is performed from the"owning node" of the cluster. The "owning node" is the clusternode that is currently running the Exchange services. In non-clusterdeployments, Double-Take would replicate from a single, physicalmachine. In a clustered deployment of Double-Take, however,replication occurs from the "owning node" and switches to anothernode should MSCS failover Exchange to another node in the cluster.Because Double-Take is tightly integrated with MSCS, this happensautomatically without administrator intervention.Other Considerations for Exchange and the rest of your environment. It is likely that new versions, service packs, hot fixes, and 3rd partyadd-ons will make protecting Exchange even more difficult in thefuture. By focusing on the Windows file system and OS, Double-Take Software will continue to provide value to customers with itseffective, yet simple, approach to Exchange protection. Thisapproach, along with the stability and scalability of the solution, is areason for the current leadership position Double-Take Softwaremaintains among Exchange protection technologies.When considering enterprise technologies, it can be difficult toselect vendors that support large areas of complex andheterogeneous organizations. When considering the variety ofapplications (such as Exchange, SQL, Oracle, and file services) andthe different versions of each of these applications in use, the task iseven more daunting. However, because Double-Take Softwarereplication technologies focus on data replication and then assist inthe pre-configuration of applications independently, the same levelof data protection for Exchange is equally viable for any otherWindows based application. More simply put, you can standardizeon one Windows availability solution, regardless of the myriad ofapplications in your environment.In the same light, while storage technologies will continue to growand change, Double-Take Software can protect any data on anyWindows server. Even if you change server manufacturers orstorage-solution vendors, as long as it is running a Windows serverOS, Double-Take Software will remain part of your solution.Double-Take Software has been protecting applications running onWindows file systems since Windows NT 3.51, and other serveroperating systems longer than that. "Business Continuity throughReplication" is the single focus of every person in our company.That focus, and the quality of our products, has helped Double-TakeSoftware forge relationships with HP, IBM, Dell, SunGard, Microsoftand probably your preferred reseller-integrator. Double-TakeSoftware has even been awarded the Advanced InfrastructureCompetency Certification for Exchange by Microsoft, proving thatour Exchange solutions are not only technically sound, but that theyhave been battle-tested by real clients in real emergencies.5Untitled Document 2006 NSI Software,Inc.All rights reserved.Double-Take ,GeoCluster and NSI are registered trademarks of NSI Software,Inc.Balance" and Double-Take for Virtual Systems" are trademarks of NSI Software,Inc.Microsoft,Windows,and theWindows logo are trademarks,or registered trademarks of Microsoft Corporation in the United States and/or other countries.All other trademarks are properties of their respective companies.are trademarksof NSI Software,Inc.Microsoft,Windows,and the Windows logo are trademarks,or registered trademarks of Microsoft Corporation in the United States and/or other countries.All other trademarks areproperties of their respective companies.Get the standard today: www.doubletake.com or 888-674-9495WHITE PAPER:PROTECTING MICROSOFT EXCHANGEAbout Double-Take Software NSI Software,Inc.(NSI ) doing business as Double-Take Software,provides the world's most relied upon solution foraccessible and affordable data protection for Microsoft Windows applications.The Double-Take product is the standard indata replication,enabling customers to protect business-critical datathat resides throughout their enterprise.With its partner programsand professional services,Double-Take delivers unparalleled dataprotection,centralized back-up,high availability,and recoverability.It's the solution of choice for thousands of customers,from SMEs tothe Fortune 500 in the banking,finance,legal services,retail,manufacturing,government,education and healthcare markets.Double-Take is an integral part of their disaster recovery,businesscontinuity and overall storage strategies.Double-Take Software isprivately held and headquartered in Southborough,MA.For moreinformation,please visit www.doubletake.com.For more information on Double-Take Softwareproducts and services please contact us.Double-Take Software Headquarters257 Turnpike RoadSouthborough,MA 01772Phone:800-775-4674Fax:201-656-2727Double-Take Software Sales8470 Allison Pointe Blvd.Suite 300Indianapolis,IN 46250Phone:800-674-9495Fax:317-598-0187Or visit us on the web at www.doubletake.com