vxlan-gpe: no data flow across compute nodes due to incorrect lisp remote-mapping

[Onong Tayeng]

The lisp remote-mapping created by agent is incorrect. The remote locator's IP address is zeroed out:

DBGvpp# sh lisp locator-set
Locator-set Locator Priority Weight
net-vpp-gpe-lset-1 1 1 1
<remote-1> 0.0.0.0 1 1
<remote-2> 0.0.0.0 1 1
DBGvpp#

This implies that no data can flow between VPP across separate compute nodes.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/747771

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/747771
Committed: https://git.openstack.org/cgit/x/networking-vpp/commit/?id=bcd6734f2394cbd8b23899ec564da99c96fe7a98
Submitter: Zuul
Branch: master

commit bcd6734f2394cbd8b23899ec564da99c96fe7a98
Author: Onong Tayeng <email address hidden>
Date: Mon Aug 24 22:07:17 2020 +0530

    vxlan-gpe: no data flow across compute nodes due to incorrect lisp remote-mapping

    The remote locator's IP address is not programmed correctly in VPP
    which implies that the l3 underlay network is effectively broken and
    no communication is possible across compute nodes.

    Refactoring included: push the construction of VPP-specific data
    structures into vpp.py and make the call parameters more neutral.

    Change-Id: Id12c6b1ce226dd65a7fe88e0d4044c44149db3c1
    Closes-Bug: #1892765