After performing a minor update to OSP - puppet runs are hanging when checking PCS status

Issue

• updated the controller ,now puppet is running a script that does not recognize the new rabbitmq-server node-attr and fails.
• The puppet manifest causing this issue is:

/etc/puppet/environments/production/modules/quickstack/manifests/pacemaker/memcached.pp

• The manifest:

$ cat memcached.pp
class quickstack::pacemaker::memcached {

  include ::memcached
  include quickstack::pacemaker::common
  class {'::quickstack::firewall::memcached':}

  Exec['wait-for-settle'] -> Exec['pcs-memcached-server-set-up-on-this-node']

  Service['memcached'] ->
  exec {"pcs-memcached-server-set-up-on-this-node":
    command => "/tmp/ha-all-in-one-util.bash update_my_node_property memcached",
  } ->
  exec {"all-memcached-nodes-are-up":
    timeout   => 3600,
    tries     => 360,
    try_sleep => 10,
    command   => "/tmp/ha-all-in-one-util.bash all_members_include memcached",
  } ->
  quickstack::pacemaker::resource::generic { 'memcached':
    clone_opts => "interleave=true",
  } ->
  Anchor['pacemaker ordering constraints begin']
}

• This manifest runs a script called ha-all-in-one-util.bash that fails with the following error:

Invalid Property: pcmk-controller2=rmq-node-attr-last-known-rabbitmq-server=rabbit@lb-backend-controller3,memcached
Invalid Property: pcmk-controller2=rmq-node-attr-last-known-rabbitmq-server=rabbit@lb-backend-controller3,memcached
Invalid Property: pcmk-controller2=rmq-node-attr-last-known-rabbitmq-server=rabbit@lb-backend-controller3,memcached
Invalid Property: pcmk-controller2=rmq-node-attr-last-known-rabbitmq-server=rabbit@lb-backend-controller3,memcached