The model.deploy unable to resolve to a revision for base [email protected]

[yhaliaw]

Description

Running the following await model.deploy("grafana-agent", "test-name", channel="latest/edge", base="[email protected]", num_units=0) results in the following error: juju.errors.JujuError: base "[email protected]/stable" is not supported, supported bases are: [email protected]

However, there should be revisions that fits the constriant, such as revision 316:

Adding the revision to the args works fine:
await model.deploy("grafana-agent", "test-name", channel="latest/edge", base="[email protected]", revision=316, num_units=0)

It would be good for the deployment be done without hardcoding the revision.

Urgency

Annoying bug in our test suite

Python-libjuju version

3.6.0.0

Juju version

3.5.4 (locally), 3.6/stable (in GitHub workflows)

Reproduce / Test

await model.deploy("grafana-agent", "test-name", channel="latest/edge", base="[email protected]", num_units=0)

[dimaqq]

A couple of notes off the top of my head:

1. base= argument is supported or not sully supported yet, please continue to use series= in the current versions of the library
2. in some cases charmhub is misleading, please report juju info <app> instead.

P.S. base= effort is tracked in #1156

[dimaqq]

Also, is this a machine charm / model or a k8s charm / model?

It's not possible to add a 24.04 machine using the library at the moment due to #1229, if that's your case, shell out to add the machine.