Ping¶
Checking the Health of your Traefik Instances
The ping
options allows you to enable the ping endpoint to check Traefik liveness.
The ping endpoint is reachable using the path /ping
and the methods GET
and HEAD
.
If the Traefik instance is alive, it returns the 200
HTTP code with the content: OK
.
Configuration Example¶
To enable the API handler:
ping: {}
[ping]
--ping=true
Configuration Options¶
The ping
option is defined in the install (static) configuration.
You can define it using the same configuration methods as Traefik.
Field | Description | Default | Required |
---|---|---|---|
ping.entryPoint |
Enables /ping on a dedicated EntryPoint. |
traefik | No |
ping.manualRouting |
Disables the default internal router in order to allow one to create a custom router for the ping@internal service when set to true . |
false | No |
ping.terminatingStatusCode |
Defines the status code for the ping handler during a graceful shut down. See more information here | 503 | No |
terminatingStatusCode
¶
During the period in which Traefik is gracefully shutting down, the ping handler
returns a 503
status code by default.
If Traefik is behind, for example a load-balancer
doing health checks (such as the Kubernetes LivenessProbe), another code might
be expected as the signal for graceful termination.
In that case, the terminatingStatusCode can be used to set the code returned by the ping
handler during termination.
ping:
terminatingStatusCode: 204
[ping]
terminatingStatusCode = 204
--ping.terminatingStatusCode=204