canonical / juju-dashboard-charm Goto Github PK

root@hazem-pfe:~# juju status
Model Controller Cloud/Region Version SLA Timestamp
controller localhost-localhost localhost/localhost 3.1.5 unsupported 14:09:06Z

App Version Status Scale Charm Channel Rev Exposed Message
controller active 1 juju-controller 3.1/stable 14 no

Unit Workload Agent Machine Public address Ports Message
controller/0* active idle 0 10.217.74.97

Machine State Address Inst id Base AZ Message
0 started 10.217.74.97 juju-4553c3-0 [email protected] Running
root@hazem-pfe:~# juju deploy juju-dashboard-k8s dashboard

Located charm "juju-dashboard-k8s" in charm-hub, revision 8
Deploying "dashboard" from charm-hub charm "juju-dashboard-k8s", revision 8 in channel stable on [email protected]/stable
ERROR Charm feature requirements cannot be met:

• charm requires feature "k8s-api" but model does not support it

Feature descriptions:

• "k8s-api": the Kubernetes API lets charms query and manipulate the state of API objects in a Kubernetes cluster

For additional

We have two versions of the juju_dashboard charm library:
https://github.com/canonical/juju-dashboard-charm/blob/main/k8s-charm/lib/charms/juju_dashboard/v0/juju_dashboard.py
https://github.com/canonical/juju-dashboard-charm/blob/main/machine-charm/lib/charms/juju_dashboard/v0/juju_dashboard.py

Both have the same LIBID, LIBAPI, and LIBPATCH, but there are differences between them. This doesn't seem like a good idea.

In fact, I don't really like the way this repo is structured at all. It's two copies of nearly the same code - and skew will arise over time as we forget to patch both of them. I'm not sure there's a way around this at the moment, though.

Consider using reify to render the template from config? It's designed to do just that :)

Would slot into https://github.com/canonical-web-and-design/juju-dashboard-charm/blob/master/src/charm.py#L79 quite easily

Juju would like to request that port be configurable for the Kubernetes charm so we can configure the dashboard proxy from Juju correctly. Currently this is hard coded to port 80.

Changes made in #16 have been packaged as a beta juju-dashboard charm.

The charm sets the IP address at which the dashboard is reachable into the relation data for the dashboard endpoint. On AWS, the cloud local address is sent. For now, we need to avoid network-get and just use unit-get-public-address until the networking stuff is sorted on the juju side.

juju bootstrap microk8s c
juju switch controller
juju deploy juju-dashboard
juju relate juju-dashboard controller

Debug logs:

unit-dashboard-0: 17:28:54 WARNING unit.dashboard/0.controller-relation-changed ERROR public address not found
unit-dashboard-0: 17:28:54 ERROR unit.dashboard/0.juju-log controller:0: Uncaught exception while in charm code:
Traceback (most recent call last):
  File "/var/lib/juju/agents/unit-dashboard-0/charm/./src/charm.py", line 127, in <module>
    main(JujuDashboardKubernetesCharm)
  File "/var/lib/juju/agents/unit-dashboard-0/charm/venv/ops/main.py", line 438, in main
    _emit_charm_event(charm, dispatcher.event_name)
  File "/var/lib/juju/agents/unit-dashboard-0/charm/venv/ops/main.py", line 150, in _emit_charm_event
    event_to_emit.emit(*args, **kwargs)
  File "/var/lib/juju/agents/unit-dashboard-0/charm/venv/ops/framework.py", line 355, in emit
    framework._emit(event)  # noqa
  File "/var/lib/juju/agents/unit-dashboard-0/charm/venv/ops/framework.py", line 856, in _emit
    self._reemit(event_path)
  File "/var/lib/juju/agents/unit-dashboard-0/charm/venv/ops/framework.py", line 931, in _reemit
    custom_handler(event)
  File "/var/lib/juju/agents/unit-dashboard-0/charm/./src/charm.py", line 55, in _on_controller_relation_changed
    requires = JujuDashReq(self, event.relation, event.app)
  File "/var/lib/juju/agents/unit-dashboard-0/charm/lib/charms/juju_dashboard/v0/juju_dashboard.py", line 58, in __init__
    ip = check_output(["unit-get", "public-address"]).decode().strip()
  File "/usr/lib/python3.10/subprocess.py", line 420, in check_output
    return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
  File "/usr/lib/python3.10/subprocess.py", line 524, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['unit-get', 'public-address']' returned non-zero exit status 1.

I suspect the problem was my commit here - this was the correct fix for the machine charm, but I probably shouldn't have changed it for k8s.