Bug 1306350

Summary: With RDO-manager, if not configured, the first nic on compute nodes gets addresses from dhcp as a default
Product: [Community] RDO Reporter: Raoul Scarazzini <rscarazz>
Component: rdo-managerAssignee: Hugh Brock <hbrock>
Status: CLOSED CURRENTRELEASE QA Contact: Shai Revivo <srevivo>
Severity: medium Docs Contact:
Priority: unspecified    
Version: trunkCC: abeekhof, chris.brown, fdinitto, mburns, mcornea, michele
Target Milestone: ---   
Target Release: trunk   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2017-06-18 06:34:36 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description Raoul Scarazzini 2016-02-10 16:09:29 UTC
Description of problem:

Deploying Mitaka with rdo-manager does not complete successfully if you don't want to use the first nic until you explicit this in the compute.yaml nic configs:

-
  type: interface
  name: nic1
  use_dhcp: false

Without this even if you have something like this in compute.yaml (That means that everything goes on the second nic):

...
...
            -
              type: ovs_bridge
              name: {get_input: bridge_name}
              use_dhcp: false
              dns_servers: {get_param: DnsServers}
              addresses:
                -
                  ip_netmask:
                    list_join:
                      - '/'
                      - - {get_param: ControlPlaneIp}
                        - {get_param: ControlPlaneSubnetCidr}
              routes:
                -
                  ip_netmask: 169.254.169.254/32
                  next_hop: {get_param: EC2MetadataIp}
                -
                  default: true
                  next_hop: {get_param: ControlPlaneDefaultRoute}
              members:
                -
                  type: interface
                  name: nic2
                  # force the MAC address of the bridge to this interface
                  primary: true
                -
                  type: vlan
...
...

Overcloud deploy will fail. This does not happen in RDO/Liberty.

Version-Release number of selected component (if applicable):

CentOS Linux release 7.2.1511 (Core)
instack-undercloud-3.0.1-dev54.el7.centos.noarch
openstack-tripleo-heat-templates-1.0.1-dev45.el7.centos.noarch

How reproducible:

Configure a network-isolation setup for an overcloud environment in which the first network card is not used for internal communication and try to deploy the overcloud.

Actual results:

Overcloud deploy fails on the compute validation.

Expected results:

Compute validation should succeed.

Additional info:

The network isolation setup that was used in this deployment is explained here https://github.com/rscarazz/openstack/blob/master/ospd-network-isolation-considerations.md

Comment 2 Christopher Brown 2017-06-17 16:52:35 UTC
This has been fixed for a while so can be closed.