dns_latency — DNS resolution latency

Free tier. Measures resolution latency for a set of names — through the operating system's resolver by default, or against explicit DNS servers to compare them. Slow DNS is a frequent cause of perceived slowness (logon time, application launch, share access).

Quick start

# probes.d/50-dns.yaml
- name: dns-checks
  type: dns_latency
  params:
    names: ["intranet.corp.lan", "www.example.com"]
    resolvers: ["10.0.0.53", "1.1.1.1"]   # optional; omit for the system resolver
    interval: 60

Parameters

Parameter	Must set	Default	Description
names	Yes	-	Names to resolve. Example: intranet.corp.lan, www.example.com
resolvers	No	-	DNS servers as ip or ip:port, each name measured against each; empty uses the system resolver. Example: 10.0.0.53, 1.1.1.1
timeout	No	5	Budget in seconds per lookup
interval	No	60	Seconds between cycles

Metrics

One series per (name, resolver) pair — resolver is system when no explicit servers are configured.

Metric	Unit	Description
senhub.dns.up	bool	Lookup answered within the timeout
senhub.dns.lookup.duration	s	Resolution time (only when up)
senhub.dns.answers	count	Addresses returned

A failing lookup is a measurement (up = 0), never a probe failure.

Metric reference

Every metric this probe can emit. Metric is the OpenTelemetry name the OTLP, Prometheus and Zabbix outputs derive theirs from. Name is what a Nagios check and the API metrics= filter match. PRTG channel is the label PRTG shows, placeholders filled from the series' tags.

Metric	Name	PRTG channel	Unit	Description
senhub.dns.up	senhub.dns.up	DNS {name} via {resolver} Up	#	1 when the lookup returned at least one answer within the timeout
senhub.dns.lookup.duration	senhub.dns.lookup.duration	DNS {name} via {resolver} Lookup Time	ms	Wall-clock resolution time
senhub.dns.answers	senhub.dns.answers	DNS {name} via {resolver} Answers	#	Number of addresses returned