Uncaught SubnetAllocationError leads to dangling networks when using auto-allocate operation

[Armando Migliaccio]

Steps to repro:

a) set up a very tiny default subnetpool
b) create plenty of autoallocated topologies for your tenants
c) see SubnetAllocationError: Failed to allocate subnet: Insufficient prefix space to allocate subnet size <bla> and a network staying behind.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/359485

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/359485
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=960eae574b75f58d489230032f084bb4c49591b5
Submitter: Jenkins
Branch: master

commit 960eae574b75f58d489230032f084bb4c49591b5
Author: Armando Migliaccio <email address hidden>
Date: Tue Aug 23 16:21:41 2016 -0700

    Catch SubnetAllocationError during auto-allocated-topology provisioning

    When uncaught, the auto_allocated_network stays behind. This patch
    ensures the error is handled properly and the network is indeed
    cleared when the subnetpool runs out of space.

    Closes-bug: #1616250

    Change-Id: Ic6e06d3e6c16a58977e84b371c3faaba66c9603b

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron 9.0.0.0b3 development milestone.