tcp_dial — TCP connect latency¶
Free tier. Measures the time for a raw TCP connect() to complete
against host:port targets. For a load-balancer VIP, a Citrix broker,
a domain controller or a fileserver, a measured dial is faster and more
dependable than an HTTP round trip.
Quick start¶
# probes.d/40-dial.yaml
- name: dial-core
type: tcp_dial
params:
targets: ["10.0.0.10:443", "dc01.lan:389", "files.lan:445"]
interval: 60
Parameters¶
| Parameter | Must set | Default | Description |
|---|---|---|---|
targets |
Yes | - | host:port pairs to dial. Example: 10.0.0.10:443, dc01.lan:389 |
timeout |
No | 5 |
Connect budget in seconds per target |
interval |
No | 60 |
Seconds between cycles |
Metrics¶
| Metric | Unit | Description |
|---|---|---|
senhub.tcpdial.up |
bool | Connect completed within the timeout |
senhub.tcpdial.duration |
s | Three-way-handshake time (emitted only when up) |
A refused or timed out target 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.tcpdial.up |
senhub.tcpdial.up |
TCP {target} Up | # | 1 when the TCP connect completed within the timeout |
senhub.tcpdial.duration |
senhub.tcpdial.duration |
TCP {target} Connect Time | ms | Time for the TCP three-way handshake to complete |