A Node is a resource that is either a physical or virtual instance of a network accessible host. Nodes have a few basic attributes but a Node’s attributes can be extended to include arbitrary named key/value pairs. Attributes typically describe the properties of a node or reflect the state of the node. One of a Node’s built in attributes is called “tags” which is a list of classifications or categories about that Node.
To add nodes to your Rundeck instances, see Configure - Resource Model Sources Configuration
Navigate to this page by clicking on the “Nodes” tab in the navigation bar.
The screenshot above shows the elements of the Nodes page user interface.
Because nodes might have many attributes, it can be helpful to use a naming convention that groups them into namespaces. The naming convention is simple:
{namespace}:{name} = {value}
When Rundeck sees an attribute with a name containing a colon it automatically groups them in the UI. For example, the screenshot below shows a node with two attributes under the “web” prefix.
The two attributes are:
This naming convention is really just a graphical treatment to reduce the amount of data being shown all at once. This convention also helps add structure to your resource model.
The node detail view contains a set of hyperlinks that automatically filter the nodes in the project. By clicking on a Node’s tags, attribute names or attribute values the list of nodes will be refined to match a more specific expression. You can delete parts of the expression from the filter form to expand the search again.
As you click search links in the node detail view, that attribute is placed in the node filter form and the attribute becomes a new column in the list view. These filtered list views are useful to create custom views about the project nodes.
The screenshot below shows a case where two nodes were found that were tagged “www” have an attribute named web:pid and with their web:state attribute value “RUNNING”.
The next screenshot shows a single node view. Notice the nodes unique name is entered in the filter form.
A node filter is a search pattern that matches a set of nodes. Rundeck node filters let you match Nodes based on a pattern like “tags” or any node attribute. You can define node filters that include and exclude nodes based on sets of patterns. While a node filter can specifically reference nodes using their unique name, filter patterns are more resilient to changes in your environment and better handle dynamically provisioned servers.
The filter expression syntax is very simple. You can find a list of nodes by entering their names, or by attribute name and value. Values can also be regular expressions.
For full syntax information see the Node Filters chapter.
If you forget the syntax, press the help button in the filter form and a dialog containing a cheat sheet will be displayed (see screenshot below).
To exclude nodes that match a pattern place a “!” (exclamation point) to negate that part of the match.
Often times you will want to go back to a set of nodes that matched a filter. You can save the filter and give it a name by pressing the “save filter” button to the right of the filter form. The example below shows the dialog to confirm saving a new filter.
The saved filters become available in the Filters menu in other parts of rundeck, like in the Nodes page, when creating Jobs or running Commands.
Often times you are browsing and filtering nodes because you want to do something with the list. The Node Actions menu contains links to
Users that don’t have sufficient privileges granted by the ACL will show that menu item disabled.
The example below shows that the user can’t create Job for those nodes.