Fax Server Redundancy – Protecting Your Document Communications

Wikipedia defines redundancy as, “the duplication of critical components of a system with the intention of increasing reliability of the system, usually in the case of a backup or fail-safe.”

See http://en.wikipedia.org/wiki/Redundancy_(engineering).  Makes sense so far, right?  We all know however, that redundant systems can increase overall system availability too, meaning users or processes can run more efficiently. Still for being such a broad topic with many variables, we do see all kinds of monikers used when it comes to this subject:   High-availability; load-balancing; fault-tolerant, disaster recovery, and so on.  From an abstract perspective all of them are in some way descriptive of redundancy, with the end result real easy to understand: No loss of data and interruptions to your business.

A business interruption can be anything; telephone or internet failures; a corrupt database; a computer virus, a bad disk drive, a failed CPU in the datacenter, or any unplanned system downtime. They can even be as far reaching as a full scale disaster; floods, power outages, hurricanes, or even terror attacks. It goes without saying, but measures must be made to ensure that all mission critical applications are in some way redundant.  This could mean anything from a simple backup to a full-blown high available, redundant system. The choice is yours, and the options are plentiful.

Yet, the biggest challenge when considering redundancy for a software application like an enterprise fax server is exactly how much redundancy do you want? After all, Open Text Fax Server, RightFax Edition has a multi-layered architecture in which various components, databases, or processes can be placed into redundant scenarios that can ensure high uptimes while preserving faxes and data from being lost. This is good. Preserving lost faxes means preserving your business after all.

Getting down to some specifics:  When architecting your fax solution to provide maximum uptime it is helpful to understand the difference between fault-tolerance and system redundancy, as both can be addressed differently and have different ramifications to a business that relies on fax. A fault-tolerant system will continue to work if a single fax server component fails. Redundancy on the other hand, allows the entire fax enterprise to continue operating if one major component of that system fails. In a redundant scenario each component of the fax server must be duplicated. Typically it is common to have two fax servers operating in tandem, so that if one server fails the other server will continue to operate.

In practical terms specific to Open Text Fax Server, RightFax Edition, there are three typical redundant scenarios used:

  1. Load-balancing and Shared Services: This is a scenario in which a fax server shares its database of users, groups, printers, etc. amongst an enterprise of multiple fax servers. Typical with any company having more than one physical location, combined with the advancements made in Fax Server architecture, less fax server resources are now needed at the remote locations, making it easier to build in redundancy while leveraging your company LAN/WAN. In the end, a proper load balancing and shared services scenario will allow Fax Server to literally share its various server services and fax images across a network. In fact, many will say that this is very similar to an active-active cluster, in which shared resources not only ensure backup, but provides processing optimization and single location to administer.  Furthermore, Fax Server is optimized for high-availability, in which the database resources are shared such that the application is providing a centralized location for all company users, groups, and other data object, it does not have to be replicated anywhere else.  Here’s an example:  A fax server in Los Angeles will be “aware” of users located in the New York office because the database is shared. There is no need to have user data replicated across the enterprise. That shared database in L.A. can obviously be placed into a database cluster for redundancy purposes. All in all, taking advantage of Open Text Fax Server’s shared service architecture will boost your system reliability and your business’s ability to run without failure.
  2. A “Cold Spare” Scenario: A cold spare configuration is intended for use in the event of a long-term system shut down, a failure, or any other system interruption that may take more time to repair. Typically this is implemented as two fax servers on two hardware servers (or virtualized) as a primary server and a secondary server. It is important to note that a cold spare is not used in production but is available so that it will expedite recovery. A cold spare system is typically stored in an isolated or remote location and is considered to be “offline”. Its purpose is to be activated in the event of a primary system failure. This is a straightforward approach and offers an affordable way to have a level of redundancy if you can tolerate some manual intervention.
  3. Active-Passive Clustering: Typically, cluster environments protect against an application/service failure, system/hardware failure, site failure and even downtimes due to planned maintenance. In the case of Open Text Fax Server where a primary fax server had a failure, the business would revert to a secondary server to continue fax processing. That is of course, if a second fax server system has been setup to function as a passive “node”. What’s nice about Fax Server’s approach to this is the “node” doesn’t have to be the entire server application. Since Fax Server connects to the telephone system, there is a way to leverage the architecture to realize the benefits of clustering. Using remote Doctransports will allow you to divide up your fax channels between one active node (a.k.a a Open Text Fax Server) and a passive node (a.k.a. a backup copy of Fax Server). You still get the same channel capacity 100% of the time, but in the event of a node failure, those fax channels simply “see” the other node and keep on processing your critical fax documents. Combine this with the shared database and services and you’re now starting to build a high available redundant system unparalleled in efficiency and effectiveness.

Lots of choices and options are available to build a redundant Open Text Fax Server, RightFax Edition system and keep your business up and running. As stated before, the options are plentiful, no matter what size your company.  Be sure to work with a Open Text Fax Server VAR or Partner who can help design a plan that meets your needs.

To learn more about Fax Server, RightFax Edition and devising redundant scenarios to protect your business, download our eBook jointly published with Windows IT Pro.

We have just published this brief overview video on the Open Text Fax Server, RightFax Edition Shared Services Module which helps with redundancy.


3 Responses to “Fax Server Redundancy – Protecting Your Document Communications”

  1. 1 John Harrison December 4, 2009 at 2:13 am

    Great post and good overview of the software alternatives to establishing a redundant architecture.

    It would be great to expand on this a little further to include a discussion on the impact for inbound faxes and number portability. While it is fairly easy to redirect outbound traffic via the the Remote Doctransports you mention above, rerouting inbound phone numbers to a live node a little harder. It really boils down to functionality that needs to be provided by the organizations telephony call control (PBX’s etc.) or by the telco routing the calls.

    Does someone have a good view of how Fax over IP architectures help achieve this goal and are there some good models for handling inbound call failover that you could share?

  2. 2 Donald Lefevre December 16, 2009 at 9:37 am

    In reply to John’s response, I would like to provide some views on how one can leverage Fax over IP architectures to achieve this goal.

    When deploying multiple fax servers in a redundant manner across a Cisco Fax over IP architecture, one can leverage network elements such as Voice Gateways, Unified CM and Gatekeepers to fail over to a backup fax server when the primary fax server is no longer available. This can be accomplished with one of the following methods:
    1. Dial-peers on a Cisco voice gateway
    2. Unified CM route group
    3. H.323 gatekeeper

    Dial-peers on a Cisco voice gateway:
    Let’s start off with the definition of a dial-peer. Wikipedia defines a dial-peer as, “an addressable call endpoint, is a device that can originate or receive a call in a telephone network” (http://en.wikipedia.org/wiki/Dial_peer). So, when multiple dial-peers are configured for the same inbound digit pattern, the voice gateway can hunt among these dial-peers until an available fax server (end point) is found.

    There are a number of conditions that can trigger a voice gateway to hunt among dial-peers for alternative fax server. Such as, setup message to the primary fax server times out and a disconnect cause code from the fax server, e.g. “user (fax server) busy”. However, for a more proactive failover approach, voice gateways can be configures to repeatedly send Internet Control Message Protocol (ICMP) ping packets to the primary fax server to confirm its availability. If ping fails, dial-peer is taken out of service automatically and an alternative dial-peer can be instructed to route the fax call to a back up fax server.

    Unified CM route group:
    In this scenario, Unified CM handles the failover to a backup fax server when the primary fax server becomes unreachable. This achieved by adding two or more fax servers to a route group within Unified CM. This fax server route group specifies a distribution algorithm that specifies how Unified CM routes calls when failover occurs. Traditionally, a “top down” distribution algorithm option is used where the primary fax server is at the top and back up fax servers subsequently follow. If the primary fax server becomes unavailable, Unified CM tries the next fax server on the list.

    Unified CM does not have proactive probing like ICMP pings to determine ahead of time the status of the primary fax server. So Unified CM hunts for alternative fax servers only upon call setup message to the primary fax server times out or a disconnect cause code from the fax server is received.

    H.323 gatekeeper:
    An H.323 gatekeeper offers lots of flexibility and options while always knowing the availability of each fax server through a registration procedure and an ongoing keepalive mechanism. The H.323 gatekeeper assigns priorities between two or more fax servers with the zone prefix command. The zone prefix with the highest priority points to the primary fax server. If this primary fax server becomes unregistered, the zone prefix with the next highest priority routes the call to an alternative fax server.

    In summary, each of these methods provides a means for the Cisco Voice Gateway, Unified CM or Gatekeeper to reroute a fax call to an alternate fax server if a failover scenario occurs.

  3. 3 jiya khan February 2, 2010 at 10:32 am

    An organization’s fax server typically performs the mission-critical task of distributing
    documents which in turn initiate, facilitate and complete business processes and ultimately
    business transactions. When fax services aren’t available, business processes and
    collaboration can come to a virtual halt.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

RSS Open Text Fax & Document Distribution Group

  • SIP Trunking: Part 2 May 14, 2010
    In part 2 of the 3-part SIP Trunking series, we go beyond the simple implementation approach of just providing an IP address and explain the additional work required to get any fax server to interoperate with the SIP Trunk.
  • SIP Trunking: Part 1 May 3, 2010
    Senior Technical Trainer Adam Friedmann discusses Session Initiation Protocol (SIP) and the benefits of SIP Trunks.
  • European Distributor Automates Order Processing with Enterprise Fax and Content Management Solution April 28, 2010
    Cito Benelux replaced manual fax and order processing with an integrated fax and content management solution to automate and optimize order processing resulting in faster, error-free processing, enhanced service and compliance, and the elimination of fax machines.
  • Open Text at Interop, Las Vegas 2010 April 25, 2010
    The Open Text Fax and Document Distribution Group participated in the InterOp 2010 show held in Las Vegas during the last week in April. The event was IT focused with companies representing areas such as computer and network monitoring, VoIP, virtualization, cloud computing, hosting, and security. We had a steady stream of traffic and discussions […]
  • Three Minutes to Install an Open Text Fax Gateway March 31, 2010
    International Technical Instructor Matt Williams, is surprised at how quickly and easily he got an Open Text Fax Gateway up and running after opening the box.

%d bloggers like this: