Description of problem: When the ODL L2 agent is down and an instance is scheduled to run on the node with this agent the instance is in 'spawning' state for a long time until it fails. The expected behavior is in the case the agent is down that the binding will fail and the instance will be rescheduled to run on a different compute node. Version-Release number of selected component (if applicable): python-networking-odl-11.0.0-0.20170804111351.a27bd55.el7ost.noarch opendaylight-6.1.0-2.el7ost.noarch How reproducible: Steps to Reproduce: 1. Block the outgoing port 6640 on the Compute node 2. Wait for the agent to be down (openstack network agent list) 3. Launch an instance and verify it's scheduled to be on the above node 4. Verify that the instance is in 'spawning' state and not rescheduled to another node Actual results: Expected results: Additional info:
Checked with python-networking-odl-11.0.1-0.20170926211325.6ceec59.el7ost.noarch. Blocked outgoing port 6640 on the compute node and waited for the agent status to be down. Then tried to launch an instance on this node and it failed as expected: {"message": "Binding failed for port c53a7d09-ef59-4298-8a73-b55f5bfa145d, please check neutron logs for more information.", "code": 500, "details": " File \"/usr/lib/python2.7/site-packages/nova/compute/manager.py\", line 1829, in _do_build_and_run_instance
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2017:3462