Cause: Previously, the Compute libguestfs did not perform tear down correctly in certain cases, such as when attempting to perform SSH key injection when no SSH keys were specified.
Consequence: libguestfs qemu instances would remain running, even after the Compute instance with which they were associated was destroyed. If one attempted to kill them, they would become zombies that required a restart of nova-compute to kill.
Fix: We now tear down libguestfs correctly.
Result: libguestfs instances no longer are left running after the build stage of the main Compute instance.