instead of printing the performance statistics for dynagroup recalculation to the RHQ Server's log, write support for exposing rich performance metrics on a groupDefinition-by-groupDefinition basis through a monitor mbean.
rev2516 - performance metrics for work done by recalculation thread for group defs / dyna groups; a new service - "RHQ Server Group Definition / DynaGroups Subsystem" - under the RHQ Server exists to wrap this functionality; a "get statistics" method exist and will retrieve metrics on a groupDefinition-by-groupDefinition basis; clear statistics will reset them to zero (as will a server restart); four measurement definition exist on the service itself to help graph overall group definition metrics over time;
Is this a first-level feature, or just something to help with debugging the dynagroup recalculation feature?
I added some dynagroups with recalc, the monitor graphs and tables appear to be accurate. The dynagroup sizes did increase when I added resources. Everything else *looks* right but for instance calc time I have no way to verify independently. rev2522
This bug was previously known as http://jira.rhq-project.org/browse/RHQ-1299