Define a liveness probe for the argo-workflows server

[Julienlin]

Is your feature request related to a problem?

We experienced a case where the argo-workflows server died and was not responding. This could have been fixed easily by a restart from initiated by Kubernetes.

Related helm chart

argo-workflows

Describe the solution you'd like

We would like to have something similar to what the argo-workflows controller has:

          livenessProbe: {{ .Values.server.livenessProbe | toYaml | nindent 12 }}

values.yaml:

  # -- Configure liveness [probe] for the server
  # @default -- See [values.yaml]
  livenessProbe:
    httpGet:
      port: 2746
      path: /
    failureThreshold: 3
    initialDelaySeconds: 90
    periodSeconds: 60
    timeoutSeconds: 30

Describe alternatives you've considered

No response

Additional context

It comes to our attention that the argo workflow server does not provide a health check endpoint (see argoproj/argo-workflows#7066). But we could use the same endpoint as the readiness probe to have something that can be used as a liveness probe.