Attached microk8s does nothing, including teardown

[Stuart Bishop]

Installed the microk8s snap, which seems to be working fine. Bootstrapped a fresh lxd controller. Added microk8s. Created a model in the microk8s cloud. All this appeared to work flawlessly. However, deploying a charm does not get past provisioning, which gets stuck allocating. There is no other feedback, and juju debug-log is completely silent. The model can be removed. However, the microk8s cloud cannot be removed with remove-k8s. 'juju list-models' shows the removed models on the microk8s cloud stuck in 'destroying' state, and 'juju debug-log' still remains completely silent.

juju bootstrap --no-gui localhost lxd \
    --config automatically-retry-hooks=false --config apt-http-proxy=http://10.0.4.1:3142/
sudo snap install microk8s
kubectl config view --raw | juju add-k8s microk8s --cluster-name=microk8s-cluster
juju add-model modelname microk8s
juju deploy $JUJU_REPOSITORY/builds/mycaascharm

[Stuart Bishop (stub)]

I believe the cause is that, per kubectl confg view --raw, microk8s is listening on the loopback address.

This was accepted as valid when running add-k8s, which arguably should be caught during validation.

The main bug I see is that there was zero feedback provided to the user. Timeouts are missing and/or errors not being bubbled up.

[Canonical Juju QA Bot (juju-qa-bot)]

This bug has not been updated in 2 years, so we're marking it Low importance. If you believe this is incorrect, please update the importance.