Recently, A customer re-ran ovn migration even thought its earlier run failed. This caused some issue with data plane too. Customer should not re-run on failure of migration rather they should move back to ml2/ovs with backup they have taken. 

We need to add this caution in "Warning" section mentioned in [1].

cc: jlibosva ykarel@redhat.com 

[1]: https://docs.redhat.com/en/documentation/red_hat_openstack_platform/17.1/html-single/migrating_to_the_ovn_mechanism_driver/index#planning-your-migration-ovs-to-ovn