In the relatively few years of its existence, email has become a mission-critical application. Loss of email access can result in lost revenue opportunities, missed contractual obligations and reduced employee productivity. Adding to this are costs related to reputation damage and loss of customer goodwill.
It's not uncommon for companies to have a single email server. An email server failure can occur at any time. When it does, users typically expect email restoration and access to their messages within seconds. Assuming that an email system disruption is possible, organizations must decide how they want an email continuity solution to perform during a failover scenario. The good news is that many options exist to address email continuity.
What users expect from email
Email users generally want seamless and full-featured use of their normal email clients during a failover. Email continuity solutions that provide only limited Microsoft Outlook features during
As Microsoft Exchange is the platform of choice for many email systems, ensure that the selected email continuity solution supports Exchange applications used by the business. These special applications must continue to function without interruption during an email server failover. This is essential to maintain end-user productivity and assure regulatory compliance.
Email recovery strategies
Email continuity solutions typically recover Exchange components at an alternate location. Once the transition has been completed, users can remotely access their accounts via the Internet or other secure network services. But even after user concerns have been addressed, returning Exchange operations to its original location is probably the biggest challenge for email recovery and continuity.
Without the proper procedures and technical support, the recovery can result in data loss, excessive bandwidth utilization, days of manual IT labor and email downtime. Several strategies can be identified to minimize the pain of an email recovery.
First is the issue of resolving divergent databases in the primary (original) and the secondary (the one that has been recovered) systems. For example, there can be un-replicated email data present on the primary Exchange server and new email data generated on the secondary Exchange server which does not exist on the primary server. During recovery, these databases need to be synchronized and reconciled with data loss or damage in either server.
Second is the need to replicate the entire email database from the recovery server to the primary server or vice versa. This process could lead to data loss and a drain on bandwidth and time. Be sure to check with email recovery vendors to see if their solutions require database replication.
Third, email continuity solutions must not have a single point of failure. Hardware, software or logical data component failures could disrupt the entire recovery. Synchronizing the mail objects on both servers using object-level replication can solve this issue.
Four steps to protect Exchange
- Protect against server failures with quality hardware and component redundancy. Obtain robust, high-quality products from reputable suppliers, perform preventive maintenance and monitor server errors. Add redundant elements such as backup power and cooling.
- Minimize storage failures with storage device redundancy and RAID. This will protect data access and data integrity from hardware failures. If using shared network storage, ensure that the networks have no single points of failure.
- Prevent network failures with redundant network paths, switches and routers. Duplicate server connections to eliminate failovers caused by the failure of a single server or network component. Ensure that physical network hardware does not share common components.
- Address physical site disruptions with data replication to another site. Two approaches can deal with site disasters, such as power outages or water damage. The first is to tightly couple redundant servers across high speed/low latency channels; this can provide zero data loss and downtime. The second is to loosely couple redundant servers over medium speed/higher latency/greater distance lines. This ensures that a remote server can be restarted by a copy of the application database, only missing the last few updates.
When planning for email recovery and continuity, be careful of generic solutions, such as disk replication or virtual machine failover. These are not Exchange-specific solutions. They may appear less costly up front, but may not provide functionality such as corruption protection during data replication or divergent database resolution on failback. Instead, consider one of the many excellent turnkey solutions that offer the requisite mix of simple installation, ease of use, no ongoing maintenance burden, advanced availability features and reliability. Be sure to check vendor references as well.
Do you have comments on this tip? Let us know. Please let others know how useful this tip was via the rating scale below.
Do you know a helpful storage tip, timesaver or workaround? Email the editors to talk about writing for SearchDisasterRecovery.com.
About this author: Paul F. Kirvan, FBCI, CBCP, CISSP, has more than 20 years experience in
business continuity management as a consultant, author and educator. He is also secretary of the
Business Continuity Institute USA Chapter.
This was first published in December 2008