Red Hat Bugzilla – Bug 535430
change dynagroup backend to use availability precompute data
Last modified: 2009-08-17 16:37:00 EDT
from http://jira.rhq-project.org/browse/RHQ-965 -- ExpressionEvaluator uses a join to the rhq_availability table where endtime is null to find the latest availability for each resource.
a few months after that was implemented, we added precomputation of the current availability for a resource. we should be using that for dynagrouping purposes, as it requires only a fraction of the cost to calculate.
rev4019 - swtich dyangorup implementation to use currentAvailability precompute column instad of hitting the rhq_availability where endtime is null condition;
Need to test this on a large inventory (100k resources) with a dyna group such as "show me all down resources" and "show me all up resources" recalculating every minute.
branches/RHQ_1_2_0_GA_CP: rev4024 - swtich dyangorup implementation to use currentAvailability precompute column instad of hitting the rhq_availability where endtime is null condition;
[16:32] <joseph> you can test it by just creating an availability based expression
[16:32] <joseph> resource.availability = UP
[16:32] <joseph> resource.availability = DOWN
[16:32] <joseph> groupby resource.availability
[16:33] <joseph> if those work correctly, then it's verified
This bug was previously known as http://jira.rhq-project.org/browse/RHQ-2126