Upgrading to Traefik V2 - error: failed to decode configuration from flags: field not found, node: kubernetes

Traefik Traefik v2 (latest)
,
1

Hi,

Fairly new to Traefik so still figuring a lot of stuff out. I'm upgrading from 1.7 to 2.6 due to upgrading EKS to 1.22. Traefik pods immediately error with:

error: failed to decode configuration from flags: field not found, node: kubernetes

We set the traefik.toml as a configmap for the containers like so:

    [global]
    
    [serversTransport]
      insecureSkipVerify = true
      maxIdleConnsPerHost = 0
    
    [entryPoints]
      [entryPoints.http]
        address = ":80"
        [entryPoints.http.proxyProtocol]
          trustedIPs = ["10.0.0.0/8"]
        [entryPoints.http.http]
      [entryPoints.monitor]
        address = ":9110"
        [entryPoints.monitor.http]
      [entryPoints.prometheus]
        address = ":9090"
        [entryPoints.prometheus.proxyProtocol]
          trustedIPs = ["10.0.0.0/8"]
        [entryPoints.prometheus.http]
      [entryPoints.public]
        address = ":83"
        [entryPoints.public.proxyProtocol]
          trustedIPs = ["10.0.0.0/8"]
        [entryPoints.public.http]
      [entryPoints.redirect]
        address = ":81"
        [entryPoints.redirect.redirect]
          regex = "^http://(.*)"
          replacement = "https://$1"
        [entryPoints.redirect.http]
    
    
        
    
    [providers]
      providersThrottleDuration = "2s"
      [providers.file]
        watch = true
    
    [metrics]
      [metrics.prometheus]
        buckets = [0.1, 0.3, 1.2, 5.0]
        entryPoint = "monitor"
    
    [log]
      level = "DEBUG"
    
    [accessLog]
      bufferingSize = 0

This config was previously working for 1.7. This config is the result of running our previous config through the Migration Tool.

Follow up question: Is it necessary to switch to using IngressRoutes and Middleware etc, or can I stick with the config I have for 1.7 (currently using regular Kubernetes Ingress with annotations).

Hopefully this all makes sense, I can post more info if needed.

Thanks

2

I think I've answered the follow up question, it seems I would be best switching to using the CRD's. I could still really do with a hand with the first error I'm seeing in the logs (this still seems to persist even after switching to CRD's)

3

solved. error caused by unnoticed flag in config

4

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.