Puppet Enterprise 2019.2 has been announced. The new version should provide business teams with new ways to automate complex workflows. The new version will also improve the management of nodes in networks.
The most important point that has been improved with the new version is the management of certain points in a network, such as switches, but also firewalls. Via the Inventory option, which is accessible from the Puppet console, installed device transport modules can be added to Public Enterprise implementations.
Once that is done, network admins can manage those modules in the same way they would normally manage agentless nodes. They can also run Puppet and certain other tasks on the listed modules. Early adopters of this new feature are strongly advised to keep the number of connected devices below 100 for the time being, in order to prevent a deterioration of the performance.
Another improvement regarding nodes is the option to install Puppet agents from the Puppet console. Also there is now the option to sort lists of node groups by name instead of just by ID, if needed.
Puppet has also added the ability to use ‘plans’. These are sets of tasks that are necessary for the automation of certain complex other tasks. The orchestrator is also equipped with a new JRuby-pool, as well as configuration options for setting the maximum number of concurrent JRuby-instances. The length of the waiting time before an instance timeout occurs can also be set.
DevClass reports that Puppet Enterprise 2019.2 works with PostgresQL 11, which means that users must upgrade external instances from that data store before upgrading the master, compilers and agents. Furthermore, users are advised by Puppet to review the accompanying update documents to avoid unexpected surprises.