docker::run fails if a docker network exists with the same name

Describe the Bug

If a docker network exists with the same name as the container created by docker::run, the following error results:

Error: Failed to apply catalog: undefined method `[]' for nil:NilClass                                                                                                  
/opt/puppetlabs/puppet/cache/lib/puppet/functions/docker_params_changed.rb:86:in `detect_changes'

Expected Behavior

docker::run should work

Steps to Reproduce

Apply

  docker_network{'prometheus':
    driver => 'bridge',
  }

  docker::run{'prometheus':
    image => 'prom/prometheus',
    ports => '9090:9090',
    restart => 'always',
    net => 'prometheus',
  }

Environment

• Version 10.2.0
• Platform Ubuntu 24.04, Puppet 8.10.0

Workaround

• Rename network or container
• Remove network manually
• Re-apply catalog

Additional Context

The issue seems to be that docker_params_changed.rb runs docker inspect and doesn't check whether it's getting results for the right thing.

param_changed = true if opts['image'] && opts['image'] != inspect_hash['Config']['Image']

^ For the network, there is no Image key.

Even if this doesn't need to get fixed, finding this on a search may help someone not waste time.

Start in docker_params_changed.rb around line 86, then reproduce the issue with the docker_network and docker::run Puppet resources shown in the report on the stated environment. Check how docker inspect results are handled when the network and container share a name. Done means docker::run applies successfully without the nil error and the existing behavior remains intact.