Charmed ingress on k8s with Traefik and Traefik-Route

The traefik charm is a charm to provide ingress to another charmed application ‘the juju way’. The idea is that if a charm integrates with traefik-k8s then you can relate the two applications and your application will receive the url at which ingress is made available.

The traefik charm supports two standardized interfaces:

  • ingress

    Using this interface, each charmed application can request a single, cluster-unique url for ingress. You can choose between a domain-name-based url ( and a path-based url (\your\parameters).

  • ingress-per-unit

    Using this interface, each charmed application can request a cluster-unique url for each existing unit. This is for applications such as prometheus, where each remote-write endpoint needs to be routed to separately, and database applications who wish to do client-side load-balancing.


The traefik route charm is a proxy charm that sits between traefik and a charm in need of ingress, and is used to provide low-level access to traefik configuration, as well as to allow per-relation configuration.

Want to have full access to all the expressive power of traefik’s routing configuration? Want to have one traefik instance, and provide domain-name-based url routing to some charms, but path-based url routing to some others? This is how you do it.

How to add ingress to your charm

Traefik owns two charm libraries to facilitate integrating with it over ingress and ingress_per_unit. At the time of writing, the most recent ingress version is v2. You can verify what the latest version for the libraries is by visiting the documentation pages on charmhub:

The following steps assume we want to use ingress. The process for ingress_per_unit is very similar.

Fetch the latest ingress library

charmcraft fetch-lib charms.traefik_k8s.v2.ingress

This will download lib/charms/traefik_k8s/v2/ The simplest way to use the library is to instantiate the IngressPerAppRequirer object from your charm’s constructor. You can immediately pass to it the host and port of the server you want to be ingressed (useful if they are static), or you can defer that decision to a later moment by using the IngressPerAppRequirer.provide_ingress_requirements API.

# src/
from charms.traefik_k8s.v2.ingress import IngressPerAppRequirer, IngressReadyEvent

... # your charm's __init__(self, ...):
        self.ingress = IngressPerAppRequirer(self, host="", port=80)
        self.framework.observe(self.ingress.ready, self._on_ingress_ready)
        self.framework.observe(self.ingress.revoked, self._on_ingress_revoked)

    def _on_ingress_ready(self, event: IngressPerAppReadyEvent):
        self.unit.status = ops.ActiveStatus(f"I have ingress at {event.url}!")

    def _on_ingress_revoked(self, _):
        self.unit.status = ops.WaitingStatus(f"I have lost my ingress URL!")

    def _foo(self):
        self.ingress.provide_ingress_requirements(host="", port=42)

IngressPerAppRequirer will take care of communicating over the ingress relation with traefik-k8s and notifying the charm whenever traefik replies with an ingress URL or that URL is revoked for some reason (e.g. the cloud admin removed the relation).

How to get the proxied endpoint exposed by traefik

You have added an ingress integration to your charm and you have deployed it alongside traefik-k8s and related them. Run the following command to get a list of the endpoints currently exposed by traefik, one for each application integrated over ingress and one for each unit related over ingress_per_unit.

juju run traefik/0 show-proxied-endpoints

These are the URLs at which your workloads are externally accessible.

Congratulations for Charmed ingress on k8s.

I think now maybe I can get access to applications without Nginx Ingress.

I’m on Microk8s/Aws and for

juju deploy traefik-k8s --to 0

the return is:

hook failed: “storage-attached”

In the log appears:

File “/var/lib/juju/agents/unit-traefik-k8s-1/charm/lib/charms/observability_libs/v0/”, line 279, in _namespace with open("/var/run/secrets/", “r”) as f: FileNotFoundError: [Errno 2] No such file or directory: ‘/var/run/secrets/’ unit-traefik-k8s-1: 13:16:49 ERROR juju.worker.uniter.operation hook “configurations-storage-attached” (via hook dispatching script: dispatch) failed: exit status 1 unit-traefik-k8s-1: 13:16:49 INFO juju.worker.uniter awaiting error resolution for “storage-attached” hook

I also tried juju deploy from zinc-k8s and it returned the same error.

I am using cStor openebs for PVC in addition to the Microk8s default volumes configuration.

What can I do to get out of this impasse, @ppasotti ?

Thanks if you can answer

Ah, just noticed the discussion is also going on here. Let’s move.

1 Like

Oi, @ppasotti

I am very grateful for your comments regarding the error I presented when deploying Traefik. You were sharp and finely tuned, and completely correct. And it made me see that I was indeed outside the K8s environment, although in the public cloud. But, I see that the volume problem was also reported in another issue. And now that I’ve corrected myself, I’m going to try to deploy it again, despite being easier with Nginx. Success, @ppasotti :clap:

Good luck deploying traefik on k8s! And thanks for the kind words :slight_smile:

1 Like