Description of problem: Agent platform having no storage node is being auto-imported into inventory Version-Release number of selected component (if applicable): RHQ 4.8 build:bf44b14 How reproducible: always Steps to Reproduce: 1. Server with rhq storage installed (ip1) 2. agent installed and connected to server (ip1) 3. agent installed and connected to server (ip2) Actual results: both ip1 and ip2 platforms inventoried bypassing discovery queue Expected results: only ip1 platform is inventoried bypassing discovery queue Additional info: ip1: 10.16.23.133 ip2: 10.16.23.140
I was unable to reproduce this using Armine's environment. Although, there was something odd to start with. I was unable to uninventory the agent platform until I uninventoried the children first. I can't explain that. But, afterwards things worked as expected. I don't think there is an issue with the autocommit as it explicitly works with only the parent platform of the StorageNode resource. Unless we can recreate the issue again, I will say that this was a transient environmental issue. But, we need to keep an eye out for anything else odd in this area, that may support evidence of some kind of issue.
Re-tested with new environment: 10.16.23.63 server 10.16.23.63, 10.16.23.106, 10.16.23.198 agents It must have been an environment issue. Marking bug as not-bug.