Appendix G. Constant Quality Monitoring - technical details

Table of Contents

The FireBrick provides constant quality monitoring. The main purpose of this is to provide a graphical representation of the performance of an interface or traffic shaper .

100 second interval statistics available graphically as svg or png and in text as csv covering at least the last 25 hours (one day)
Loss latency stats where available including minimum, average, and maximum latency for the 100 second sample, and percentage packet loss.
Throughput stats where available (e.g. interfaces, shapers ) including average tx and rx rate for 100 second sample

Graphs can be loss/latency or throughput of both. A ping only system would only have loss/latency. An interface or shaper normally has only throughput data.

G.1. Tx/Rx direction

A graph shows information about two directions, tx and rx. In many cases this is simple - a graph attached to an interface has rx for traffic coming in to the FireBrick, and tx is for traffic leaving.

However, a graph linked to a firewall rule is more complex. This is explained in the firewall rules with set-graph and set-reverse-graph settings. For a firewall rule session being graphed the rx and tx relate to the direction the session is set up. You can deliberately reverse this using set-reverse-graph.

The reason this may seem complex is when making a firewall rule that has, for example, a matching of ip for an IP you want to monitor, and sets a graph. Sessions started to the IP address will have tx and rx reversed compared to sessions started from the IP address. The solution is two rules, one with target-ip and set-graph, and a separate one with source-ip and set-reverse-graph (which can be the same graph). This will then result in consistent tx and rx relating to traffic directed to or from the IP address.


F.6. Advanced commands		G.2. Access to graphs and csvs