Nfdump Exporter

This is a prototype exporter for nfdump. It exposes metrics processed by the Prometheus monitoring system.

It's purpose is to play and experiment with nfdump netflow data and Promtheus/Grafana to build a new graphical UI as a repacement for aging NfSen.

This experimental exporter exposes counters for flows/packets and bytes per protocol (tcp/udp/icmp/other) and the source identifier from the nfcapd collector. (currently hardwired "live"). Multiple collectors (ident) with multiple exporters each may send metrics to the exporter.

Metrics:

  namespace = "nfsen"
	uptime = prometheus.NewDesc(
		prometheus.BuildFQName(namespace, "collector", "uptime"),
		"nfsen uptime.",
		[]string{"version"}, nil,
	)
	flowsReceived = prometheus.NewDesc(
		prometheus.BuildFQName(namespace, "collector", "flows"),
		"How many flows have been received (per ident and protocol (tcp/udp/icmp/other)).",
		[]string{"ident", "exporter", "proto"}, nil,
	)
	packetsReceived = prometheus.NewDesc(
		prometheus.BuildFQName(namespace, "collector", "packets"),
		"How many packets have been received (per ident and protocol) (tcp/udp/icmp/other).",
		[]string{"ident", "exporter", "proto"}, nil,
	)
	bytesReceived = prometheus.NewDesc(
		prometheus.BuildFQName(namespace, "collector", "bytes"),
		"How many bytes have been received (per ident and protocol) (tcp/udp/icmp/other).",
		[]string{"ident", "exporter", "proto"}, nil,
	)

Usage:

Usage of ./nfsen_exporter:
  -UNIX socket string
    	Path for nfcapd collectors to connect (default "/tmp/nfsen.sock")
  -listen string
    	Address to listen on for telemetry (default ":9141")
  -metrics URI string
    	Path under which to expose metrics (default "/metrics")

The nfsen_exporter listens on a UNIX socket for statistics sent by the nfcapd collector.

Add this to prometheus.yml:

  - job_name: "nfsen"

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.

    static_configs:
      - targets: ["localhost:9141"]

Nfdump

The metric export is integrated in nfdump 1.7-beta

In order not to pollute an existing nfdump netflow installation, forward the traffic from an existing collector. Add: -R 127.0.0.1/9999 to the argument list and setup the new collector. You may also send it to another host, which runs also Prometheus for example.

Build nfdump 1.7-beta:

git clone -b unicorn https://github.com/phaag/nfdump.git nfdump.unicorn

Build nfdump with sh bootstrap.sh; ./configure but do not run make install, as it would replace your existing installation. Create a tmp flow dir and run the collector from the src directory. For example:

./nfcapd -l <tmpflows> -S2 -y -p 9999 -m <metric socket>

If adding -m <metric socket> nfcapd exports the internal statistics every 5s the the exporter.

Please use https://github.com/phaag/nfdump_exporter/discussions/ for discussions.

Note:

Only the statistics values are exposed and not the netflow records itself.