Tinnac is a merger of two separate trains of thought, one tracking when something goes critical, and the other tracking general wellbeing of a system.

Nagios is an outstanding example of the former, and Cacti is an outstanding example of the latter. Unfortunately Nagios doesn't graph or track longterm performance data very well, and Cacti doesn't alert nor have escalations, this is where Tinnac comes in. By taking the best ideas of both, eliminating the HORRIBLE interfaces that have plagued them both for years, and including some outside resources neither had previously included, Tinnac hopes to fully replace Nagios and Cacti with what is simply a better product.

Tinnac uses similar concepts and utilities well known among its ideological forefathers such as RRD, RRDTool, even going so far as to using a forked version of NRPE (although is mostly compatible with existing NRPE installations) and can run most existing Nagios checks (more by their own design than my ours). Topping this off is a very much improved interface allowing easy configuration of many systems, including bulk changes, something many large scale systems administrators have loathed about Nagios and Cacti. Backing this up Tinnac uses a pair of Databases, one for global configuration using Mysql which allows easy integration for Outside projects, and a BDB database which is generated from the data in the Mysql Master, for use when the Mysql database is down. During such as event performance data is stored locally and pushed up to the Mysql database when it is available. This allows Tinnac to continue functioning properly without a database, but allows all the wonderful ease of use a database provides, no more flat files, everything can be edited through the web interface, or through a handful of CLI utilities. Using this design allows distributed monitoring much easier, basically turning the monolith the network monitor system used to be into a band of managers, each controlling their own realm, and allowing for fail over redundancy.