TCP Health
Description
Collects status of a TCP server.
Parameters
| Parameter | Description | Type | Default |
|---|---|---|---|
| name | Alias to give to the TCP Server | str | N/A (Mandatory) |
| host | Name of the target host or IP address to check | str | "127.0.0.1" (Optional) |
| port | TCP port where to perform the connection check. | int(>0 <65536) | 80 (Optional) |
Schema
{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "TCP Health",
"description": "Status of a TCP server",
"properties": {
"host": {
"description": "Server hostname",
"type": "string"
},
"port": {
"description": "Port number",
"type": "integer",
"minimum": 1,
"maximum": 65536
},
"server_name": {
"description": "Server alias, from the 'name' parameter",
"type": "string"
},
"active": {
"description": "Whether the TCP connection check succeeded",
"type": "boolean"
}
},
"type": "object"
}
tcp_health.json (the schema file this block is copied from) is wrong twice: active
— the boolean connection-health result collect() actually writes — isn't declared as
a property at all, and server_name's own description says "Time the system has been
up", copy-pasted from uptime.json. Corrected above; the schema file itself still has
both bugs.
Configuration
...
tcp_health:
plugin: "dc_measurements/TCPHealth"
topic_output: "/dc/measurement/rustfs_health"
group_key: "rustfs_health"
host: "127.0.0.1"
port: 9000
name: "rustfs_api"
Example output
{
"active": true,
"flattened": false,
"host": "127.0.0.1",
"name": "tcp_health",
"nested": false,
"port": 9000,
"run_id": "169",
"server_name": "test_service"
}