LUALoriotPro WEB site

LoriotPro scripting documentation
Extensions for LUA language

Table of contents Search function

Print the current page Mail this  link UTUBE Channel version française

Redundancy and degraded mode

LoriotPro™ V7 Broadcast Edition provides the capability to collect and correlate thousand of performance indicators, SNMP but not only, in distributed mode. With this innovative technology it is possible to have a permanent redundancy and real-time monitoring solution with a 100% uptime.

An internal database of monitored data (GLOBAL OBJECTS) for this purpose is distributed between two or more LoriotPro. Each of them takes care of part of the collects (SNMP and others) and correlations of thousands of state and performance indicators available in your infrastructure. The synchronization of the GLOBAL OBJECTS database is permanent and thus, guaranties the uniqueness of the values between each LoriotPro.

If we consider 2 LoriotPro in our case, in Normal state, both of them are always active to ensure their share of the collects from the systems and equipments providing a controlled load distribution.

With a continuous communication of heartbeat type between them, LoriotPro can detect the absence of one of its peer and take over its SNMP data gathering. Systems must be designed accordingly and have the necessary throughput to ensure the full work alone naturally.

By default, the database collections to realize is the same on both systems, but for each collect a parameter indicates the system of choice (Main Collector) . The main collectors are defined for each object and can be changed on the fly and during operation.

The dashboards and network maps defined in LoriotPro (Active View) use the Global Objects database to update their color status or any other display characteristics and do not take care of how the SNMP data are collected.

The diagram below is an example of operation in Normal mode.

GO_Redundancy_EN.png

In addition and to simplify configuration, a replication mechanism check that the scripts used for the collects are identical on all LoriotPro.

If one of the LoriotPro in a peer relation stops to send the heartbeat we jump in degraded mode, the other one  start the gathering of all the Global Object that are normally gathered by the defective LoriotPro.

Note: The solution is viable if the IP communications infrastructure is guaranteed. Possible dual path IP communication between LoriotPro system (heartbeat) is especially critical and should as far as possible be carried between LoriotPro without intermediate active devices.  Crossed Ethernet point- to-point copper or fiber connections are ideal.

The diagram below is an example of operation in Degraded mode.

GO_Redundancy_Backup_EN.png

In summary, the real time distribution of a single database of GLOBAL OBJECTS over multiple LoriotPro provides a good level of redundancy and can be used in normal mode for distributing the load over different sites and network links.

 



www.loriotpro.com