The basic format is a sequence of
attributename: value pairs to select nodes which match. You can negate a match by using
!attributename: value. If you do not specify an attribute name, then the
nodename is matched against the bare value.
value can be:
tagsattribute, a set of values, results in all nodes with a set of tags that match.
+-separated list of values to AND together
a+b(match node with both tags
a+b,x(match node with both tags
b, or tag
The result is all nodes which match all of the clauses.
You can include multiple attribute name/value pairs with the same attribute name, in which case it effectively concatenates all of the values into a comma-separated list, which will not use regular-expression matching for any of those values.
name: web1 name: web2
Is the same as:
This does not use regular-expression matching. If you want to combine multiple regular expressions, refer to the regular-expression syntax. For example, you could use
anode.*|bnode.* to match all nodes starting with “anode” or “bnode”.
Match a set of nodes by name, includes all the node names listed:
Include mynode and exclude mynode2:
mynode1 !nodename: mynode2
Include nodes with both of the tags
prod or either of the given hostnames:
tags: www+prod hostname: dev1.example.com,dev2.example.com