Metrics monitoring

Available since Access Gateway version 2022.2.3

Access Gateway provides a set of monitoring metrics which can be accessed by systems such as Prometheus, and others.
In general metrics based monitoring includes:

  • Metrics collector - Access Gateway implements a custom collector which tracks metrics such as Swap, CPU, Disk, I/O and much more.
  • Publication format - Access Gateway provides metrics in the Open Metrics format.
  • Transport - Access Gateway exposed metrics using either REST and a well defined endpoint, or a push mechanisms.

In general the architecture resembles:

Architecture components include:

  • One or more instances of Access Gateway.
  • An instance of Prometheus.
  • A set of configuration settings defining metrics gathering interval, allowed IP addresses and enabling or disabling metrics gathering.
  • A set of protected web resources.

Before you begin

Ensure that:

  • Port 8889 is open between the monitoring implementation and Access Gateway nodes.

Typical implementation taskflow

Task

Description

Open port 8889 Ensure port 8889 is accessible from the monitoring solution and all Access Gateway nodes. See Prerequisites for deploying Access Gateway for more information about all ports and other prerequisites for deploying Access Gateway.
Configure metrics monitoring Configurations Management console metrics monitoring.
Configuration typically includes enabling metrics based monitoring and specifying allowed target hosts.
Test Metrics monitoring using ReST Once configured, using cURL or a similar tool, test ReST calls against each configured Access Gateway instance.
Test against each Access Gateway cluster member and the admin node.
Configure clients Configure a pull based client such as Prometheus to use the metrics monitoring to monitor all cluster members.
Examine supported metrics Examine the set of supported metrics and configure the monitoring client accordingly. See Supported metrics for a complete list of all metrics, their types and descriptions.