Bug 1598335

The issue still not solved even today. 

packstack adds an IP address to br-ex, but it does not survives a reboot.

+++ This bug was initially created as a clone of Bug #1430181 +++

Description of problem:

A Packstack controller should survive a reboot. However, after a reboot it seems that br-ex is not correctly set up, so that instances can't be reached via floating IP.

Version-Release number of selected component (if applicable):
Name        : centos-release-openstack-ocata
Arch        : noarch
Version     : 1
Release     : 1.el7

How reproducible:
Easy

Steps to Reproduce:

1. Follow the Packstack quickstart and provision an all-in-one cloud: 
       packstack --allinone --default-password=xyz -d

2. After provisioning, run ip route and ip address. 
   A route like this should exist:
       172.24.4.0/24 dev br-ex  proto kernel  scope link  src 172.24.4.1
   
   and br-ex should look like this:
       4: br-ex: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state   UNKNOWN qlen 1000
           link/ether 5a:9d:96:0b:ed:48 brd ff:ff:ff:ff:ff:ff
           inet 172.24.4.1/24 scope global br-ex
              valid_lft forever preferred_lft forever

3. Reboot the controller. 

Actual results:

After rebooting:
   - the above route has disappeared
   - br-ex has no IP address and is down

Expected results:
Route and br-ex are set up in the same way as before the reboot

Additional info:
https://ask.openstack.org/en/question/103802/packstack-route-to-external-network-disappeared-after-rebooting-controller/
The problem can be fixed by manually setting the br-ex IP address.