Puppet Enterprise
(PE) provides a web console which can be used for different purposes:
-
Classify nodes in the Node Classifier, pinning single nodes or setting matching rules to assign a node to a
Node groups
and defining for eachNode Group
the list of classes and parameters to include. Nodes classification via the console is not mandatory, and actually in our base setup it's not used. -
Check and review the
Puppet run
reports for all our nodes, having insights on what has changed in the infrastructure. -
Have an overview on the healthiness of the infrastructure and insights on what's failing.
-
Manage
Puppet job tasks
. -
Have an inventory of the facts of all the managed systems.
The PE console
provides direct access to all the data generated by Puppet
and stored on PuppetDB
.
One of the most useful features of the PE console
is the Dashboard
, here we have an immediate overview on the status and the health of the infrastructure.
We can see the nodes where Puppet
is running for real, the ones where it's running in noop mode (Puppet runs
, shows the changes it would do, but it doesn't actually perform any modification on the system), and the ones which are not reporting (Puppet
has not run on the node, or has not reported back to the server, for more than one hour (default setting)).
We have also the list of the latest reports, where we can see the total number of managed resources, the number or resources changed for a change in our code (intentional changes), which ones were changed because Puppet
brought resource back to the desired state, after some local manual modification (corrective changes), the resources which failed and the total time spent by the Puppet Server
to compile the catalog (configuration retrieval).
Clicking on a report date, it's possible to see the full details of the relevant Puppet
transaction with all the managed resources and the occurred events, with the possibility to drill down to the specific line of code where the failure occurred.
Clicking on a node name, we can see the detail page for that node, here, on the different tabs, we can access to a lot of information about the node:
- The full list of the node's facts
- All the saved reports and events for the node
- The list of classes and parameters which are used on the node (NOTE: only the ones deriving from the classification on
PE console
are included, classes included inPuppet code
, on viaHiera
, are not listed here) - A graph with al the relationships among the node's resources