To add more context, the gate-tempest-devstack-neutron is showing two failure modes: 1) behaviour which is exactly about this bug - as it can be seen in http://logs.openstack.org/91/48591/2/gate/gate-tempest-devstack-vm-neutron/4e2582a/ - In this case the timeout, being caused by agent RPC calls, can happen at anytime - not just a VM boot, and indeed we have failures in tempest.api.network tests 2) behaviour without any related traceback in neutron (the ones I see are related to negative tests IMHO), as it can be seen in http://logs.openstack.org/46/47546/2/gate/gate-tempest-devstack-vm-neutron/c8da791/
Hopeefully removing vpn from the gate will reduce the occurency of #1; however I am more concerned about manifestations of type #2, since they are more frequent in the tests I have been doing on my machines.
To add more context, the gate-tempest- devstack- neutron is showing two failure modes: logs.openstack. org/91/ 48591/2/ gate/gate- tempest- devstack- vm-neutron/ 4e2582a/ - In this case the timeout, being caused by agent RPC calls, can happen at anytime - not just a VM boot, and indeed we have failures in tempest.api.network tests logs.openstack. org/46/ 47546/2/ gate/gate- tempest- devstack- vm-neutron/ c8da791/
1) behaviour which is exactly about this bug - as it can be seen in http://
2) behaviour without any related traceback in neutron (the ones I see are related to negative tests IMHO), as it can be seen in http://
Hopeefully removing vpn from the gate will reduce the occurency of #1; however I am more concerned about manifestations of type #2, since they are more frequent in the tests I have been doing on my machines.