During upgrade plan do not stop the pe-postgresql service on the Primary Master

[mristok]

Describe the Bug

Currently the peadm::upgrade plan stops the 'pe-postgresql' service on the Primary Master before executing the puppet-enterprise-installer. However, during the upgrade one of the first steps is to backup the Postgresql instance which requires that the service is running.

# Shut down pe-* services on the master. Only shutting down the ones
# that have failover pairs on the master replica.
['pe-console-services', 'pe-nginx', 'pe-puppetserver', 'pe-puppetdb', 'pe-postgresql'].each |$service| {
  run_task('service', $master_target,
    action => 'stop',
    name   => $service,
  )
}

Expected Behavior

Remove 'pe-postgresql' from the list of services that are being stopped before the upgrade is initiated.

Steps to Reproduce

Steps to reproduce the behavior:

1. Start with a deployed instance of PE, running at >=2019.0
2. Execute upgrade plan
3. Failure during puppet-enterprise-installer task, due to pe-postgresql service being stopped

run_task('peadm::pe_install', $master_target,
  tarball => $upload_tarball_path,
)

Environment

• Version = 2019.0
• Platform Ubuntu 18.04

Additional Context

[reidmv]

I believe this was resolved by #56.