This is a small utility with no pretentions whatsoever trying to solve very simply the common question how do I extract the many monitoring items of MongoDB toward my monitoring or capacity planning system, or simply store them myself?

When all modules are enabled, it provides:

• Data collection and historization inside a MongoDB database of a configurable set of data points, as well as of the traced queries.
• An optional web application (self hosted, can be put behind a reverse proxy) for first level analysis providing:
  • a web dashboard giving the instant health of you DB as well as the ability to query the history
  • a set of very simple web services (JSON or XML) to enable you to retrieve the data from another system (Nagios, RRDB, ...)
• A Zabbix agent, and therefore can be queried directly from Zabbix (including auto discovery) - templates are provided.

It is extremely easy to use:

• Runs on Windows and Linux.
• Packaged as RPM (Linux) and soon as Chocolatey (Windows). It even includes the runtime it uses (dotnet core) to ease installation on all platforms.
• Compatible with both single instances and replica sets. New replicas (and removed replica) are discovered without restart or configuration change.
• Compatible with password database authentication, or no authentication.
• Can be deployed either on the DB server itself, or on another machine to create a centralized monitoring system for multiple instances (as it uses standard MongoDB connections).
• Only collects the metrics you need - it is fully customizable, and is provided with a default configuration which should suit most users.
• All three modules are optional - once again, just select what you need.

See the installation documentation. Note that a single installation can monitor multiple MongoDB instances, so a single centralized instance may be enough for all your servers.

After the install, if you want to use the Zabbix MongoDB agent, follow this procedure.

See this document.