Failure and Recovery in Network Video Recording

Stay Informed

When it comes to video surveillance, nothing is worse than having an incident, thinking you have it on camera via your network video recorder (NVR), and then finding out you had an NVR failure during that time… and you capture nothing!Murphy’s Law can rise up and bite us at any time (if things can go wrong, they probably will). The good news is that this can be easily avoided with failure/recovery infrastructure within your surveillance environment. During video surveillance system operations, many unexpected problems can occur in an NVR, such as failure of the network, power supply or hard disk drive, interrupting video recording and causing video information to be lost.

A single camera failure impacts video over a single channel, but what if the video storage device (NVR) fails? This can mean a loss of video data across the entire network supported by this storage device. This situation can be devastating. Of course, if your video feeds are monitored by individuals in real time, when a disruption occurs, it can be dealt with in real time. But what happens if real-time video monitoring is not happening? In this case, critical video feeds can be lost with devastating consequences. Depending upon how long it takes to recognize this failure could mean the difference between a minor blip or a major catastrophe.

There are many critical infrastructures that require 24/7/365 video monitoring such as banks, airports, oil and gas utilities, and electrical and nuclear power plants. In the event of a video storage system failure, these facilities can have serious consequences such as substantial financial fines from regulators for lack of regulatory compliance.

Any quality video management system features a failure/recovery function to minimize data loss by switching to a different NVR or server that can offer system redundancy, even if the recording on the initial NVR is interrupted by its total failure. Another way to phrase this is failover technology.

Understanding Failover Technology

Failover technology is one solution for dealing with video disruption that can happen during system operations. Another way to look at it is as a back-up system that can automatically take over the moment disruption occurs from either regular maintenance or a system malfunction. This failover technology approach is frequently used for servers, network systems, and databases.

A failover feature consists of two systems: a primary system that is regularly used and a back-up system that serves as a secondary system in case the primary system fails. The failover configuration is determined by the number of primary systems that can be backed up and the number of available secondary systems.

A failover system in the N+1 configuration would consist of an unlimited number of primary systems and a single or one secondary system. In case of a failure, the secondary system can only fill in for one primary system. It therefore cannot provide 100 percent backup operation if two or more primary systems fail.

As an example, if an NVR system supports 10+2 failover, it means that two secondary systems can serve as the backup for up to ten primary systems. An NVR capable of N+M failover system means that there are no restrictions on the number of primary and secondary systems.

N+1 Failover Systems: A Closer Look

One secondary NVR stands by at all times. If the main NVR fails, then the secondary NVR is activated. Once the main NVR is recovered, the status of the secondary NVR changes back from active to standby.

Within the NVR, the communication between processes is typically handled via Inter-Process Communication (IPC), and failover-related communication between NVRs is handled via net socket. Failover manager and failover agent processes are, in most cases, included in the main process of the NVR. The failover agents of each NVR exchange the status of NVR and control each other via the failover manager.

If a normal NVR fails, then the standby NVR's failover agent sends various system information (e.g. network camera information) registered from the normal NVR to the standby NVR. The standby NVR then provides the same services as the normal NVR. The configuration data is transferred between failover agents on the normal and standby NVRs during normal operations.

The system status is categorized internally into four states, each with designated roles such as diagnosing a failure and sending failure details. Failback is a function that sends the video data stored on the standby NVR to the normal NVR after the failure has been resolved through the failover process. The failback also restores the original NVR back into normal operations and the standby NVR back into standby mode.

Don’t Miss a Minute of Surveillance With Failover Technology

When thinking about video surveillance, Murphy’s Law can really be a challenge. But by implementing a solid failover solution infrastructure, you can be sure that what you need to record will be recorded when it needs to be recorded.

Stated another way, within any good NVR, failover function relieves concerns regarding loss of data in case of failures and strengthens the reliability of the recorded data. Another advantage of failover functionality is that a failover system can be set up simply by deploying an additional NVR or PC as a secondary system, without requiring full N+N redundancy allowing for cost savings.

New call-to-action


The opinions expressed in this piece are solely Hanwha Techwin's. They do not necessarily represent WESCO’s views.