Redundancy

We have been receiving many questions regarding redundant systems. This is the official IPitomy application for having redundancy. It is worth considering the massive complications involved in over simplifying redundancy. To truly be redundant, you simply require two of everything. That means two PBX systems, two carriers for PRI and Two PRI’s, Two carriers for SIP and Two separate SIP providers and Two Internet service providers and two separate connections.

Very few people take the time to consider all of the options. The company wishing to have redundancy needs to have a budget and a policy that governs their individual requirements. Before proceeding down the path of selling a redundant system, it is wise to ask the customer what their budget is, what their disaster recovery policy is. If they don’t have a policy or budget, then identifying the customers concerns will become the driving force in dictating what all comprises their redundant PBX.

Hardware Failure

RAID hard drives are the lowest cost and most effective

RAID and a redundant power supply defend against hard drive and power supply failure

A secondary server defends against complete system failure

PSTN Lines

Having Redundancy in the PSTN lines that guards against line failure may be a consideration. Adding lines to the secondary and primary servers are considerations that defend against line failure, especially if they are from different carriers. SIP trunks can add an additional element of redundancy at a very low cost and will resolve the line failure consideration. Multiple SIP carriers can reduce the risk as well.

Internet

Having more than one internet service provider can also protect against loss of Internet on a single carrier.

There are many different options, included in the diagram.

The redundant application requires two fully Licensed systems. One could be in the cloud, or could be local. In the case of companies using a PRI or analog lines as part of their redundant application, it is important to keep in mind that if the backup system is in the cloud, only SIP trunks apply. The PRI or analog lines may be forwarded to the SIP trunks, but there is no way to reasonably have those lines on a cloud based system.

The telephones require programming of each set to have a secondary server. This will ensure that the phones will automatically register to the secondary system in case the primary system is not available. When the primary comes back on line, the phones will automatically switch over to the primary.

It is recommended to include voice mail to email always at part of the application as the possibility exists that the voice mails in the primary server and the secondary server will be out of synch. In order to keep old voicemails and get new ones, it is essential to use voice mail to email.

The PRI has a couple of options. The PRI can be manually moved over to the secondary system. This is good because then all of the DID’s will work properly. If the secondary system has a dedicated SIP trunk, this is good so that the period of time it takes to move the PRI cable over to the secondary server, 911 will be accessible. The application can also include a second dedicated PRI in the secondary system. If the secondary PRI is from a different provider, the redundancy is obviously enhanced. Consideration for forwarding must be part of the planning process when setting up the corporate disaster recovery plan.

Having SIP trunks in the primary server can provide a good option as they can be set to automatically switch to the secondary by the provider when the primary is not available. Having two SIP providers and two Internet service providers also adds to the overall redundancy. When using multiple sip providers and Internet providers, the corporate disaster recovery policy needs to take into account the ability of disparate providers to forward DID’s and adjust the call routing to utilize automated attendants and live answering to accommodate inbound calls.