There are a couple of ways that a tenant can setup their subnet to cause the DHCP agent to go into a failure loop when it tries to create a port. This creates extra load on the agent that makes its response to other changes slower.
There are two fixes that should be back-ported to all of the mos-neutron branches.
There are a couple of ways that a tenant can setup their subnet to cause the DHCP agent to go into a failure loop when it tries to create a port. This creates extra load on the agent that makes its response to other changes slower.
There are two fixes that should be back-ported to all of the mos-neutron branches.
The first has merged so it's ready to be back-ported: https:/ /review. openstack. org/#/c/ 177187/
The second is still under review, but should be ready in the next couple of days: https:/ /review. openstack. org/#/c/ 157697/