Hello,
We are running Traefik in Kubernetes (installed via Helm).
When we create an Ingress object in K8S, the event is read almost immediately and the Host is updated on the object. However, the route in Traefik is not configured for up-to 3 minutes, meaning the URL responds with 404 until then.
I did see an error similar to this float by at one point but it wasn’t consistently happening, so I don’t know if that is a red herring. Other than that there are no obvious errors at all - enabling debug logging didn’t reveal why the route provisioning was taking so long.
We tried adjusting these two settings upwards but it didn’t make any obvious difference:
--providers.kubernetesingress.throttleduration=1s
--providers.providersthrottleduration=1s
I don’t know anything much about the internals of Traefik. Has anyone seen this before or know where the problem might be happening within Traefik?
Thanks
James