Re-deploying charm results in `caas-unit-init failed` error

While working on my charm, I encountered this error and wasn’t sure yet where to file the bug so I’m asking this question here in the meantime.

When I remove my properly-running application using the following command:

$ juju remove-application --destroy-storage --force prometheus
removing application prometheus
- will remove storage database/0

And then re-deploy it with:

$ juju deploy . --resource prometheus-image=prom/prometheus:latest
Deploying charm "local:kubernetes/prometheus-2".

juju status returns with:

Model  Controller  Cloud/Region        Version  SLA          Timestamp
lma    mk8s        microk8s/localhost  2.7.6    unsupported  10:39:52Z

App         Version  Status  Scale  Charm       Store  Rev  OS          Address         Notes
prometheus           active      1  prometheus  local    2  kubernetes  10.152.183.206

Unit           Workload     Agent  Address     Ports     Message
prometheus/1*  maintenance  idle   10.1.95.54  9090/TCP  Configuring pod

But then prometheus/1 gets stuck in that status indefinitely. Looking at the logs, I see the following error:

application-prometheus: 10:34:30 ERROR juju.worker.dependency "unit-init-worker" manifold worker returned unexpected error: initializing unit "prometheus/1": caas-unit-init failed: : command terminated with exit code 137

Currently the workaround I’m seeing that works is to completely destroy the model and its storage as follows:

$ juju destroy-model --destroy-storage lma

What steps can I take to further investigate and what’s the right place to file a bug if ever? I’m guessing this should be filed as a Juju bug but wanted to check here first to be sure.