Applies to SUSE OpenStack Cloud 9

8 Configuration Processor Information Files #

In addition to producing all of the data needed to deploy and configure the cloud, the configuration processor also creates a number of information files that provide details of the resulting configuration.

These files can be found in ~/openstack/my_cloud/info after the first configuration processor run. This directory is also rebuilt each time the Configuration Processor is run.

Most of the files are in YAML format, allowing them to be used in further automation tasks if required.

File	Provides details of
`address_info.yml`	IP address assignments on each network. See Section 8.1, “address_info.yml”
`firewall_info.yml`	All ports that are open on each network by the firewall configuration. Can be used if you want to configure an additional firewall in front of the API network, for example. See Section 8.2, “firewall_info.yml”
`route_info.yml`	Routes that need to be configured between networks. See Section 8.3, “route_info.yml”
`server_info.yml`	How servers have been allocated, including their network configuration. Allows details of a server to be found from its ID. See Section 8.4, “server_info.yml”
`service_info.yml`	Details of where components of each service are deployed. See Section 8.5, “service_info.yml”
`control_plane_topology.yml`	Details the structure of the cloud from the perspective of each control-plane. See Section 8.6, “control_plane_topology.yml”
`network_topology.yml`	Details the structure of the cloud from the perspective of each control-plane. See Section 8.7, “network_topology.yml”
`region_topology.yml`	Details the structure of the cloud from the perspective of each region. See Section 8.8, “region_topology.yml”
`service_topology.yml`	Details the structure of the cloud from the perspective of each service. See Section 8.9, “service_topology.yml”
`private_data_metadata_ccp.yml`	Details the secrets that are generated by the configuration processor – the names of the secrets, along with the service(s) that use each secret and a list of the clusters on which the service that consumes the secret is deployed. See Section 8.10, “private_data_metadata_ccp.yml”
`password_change.yml`	Details the secrets that have been changed by the configuration processor – information for each secret is the same as for `private_data_metadata_ccp.yml`. See Section 8.11, “password_change.yml”
`explain.txt`	An explanation of the decisions the configuration processor has made when allocating servers and networks. See Section 8.12, “explain.txt”
`CloudDiagram.txt`	A pictorial representation of the cloud. See Section 8.13, “CloudDiagram.txt”

The examples are taken from the entry-scale-kvm example configuration.

8.1 address_info.yml #

This file provides details of all the IP addresses allocated by the Configuration Processor:

  NETWORK GROUPS
     LIST OF NETWORKS
        IP ADDRESS
           LIST OF ALIASES

Example:

  EXTERNAL-API:
     EXTERNAL-API-NET:
        10.0.1.2:
           - ardana-cp1-c1-m1-extapi
        10.0.1.3:
           - ardana-cp1-c1-m2-extapi
        10.0.1.4:
           - ardana-cp1-c1-m3-extapi
        10.0.1.5:
           - ardana-cp1-vip-public-SWF-PRX-extapi
           - ardana-cp1-vip-public-FRE-API-extapi
           - ardana-cp1-vip-public-GLA-API-extapi
           - ardana-cp1-vip-public-HEA-ACW-extapi
           - ardana-cp1-vip-public-HEA-ACF-extapi
           - ardana-cp1-vip-public-NEU-SVR-extapi
           - ardana-cp1-vip-public-KEY-API-extapi
           - ardana-cp1-vip-public-MON-API-extapi
           - ardana-cp1-vip-public-HEA-API-extapi
           - ardana-cp1-vip-public-NOV-API-extapi
           - ardana-cp1-vip-public-CND-API-extapi
           - ardana-cp1-vip-public-CEI-API-extapi
           - ardana-cp1-vip-public-SHP-API-extapi
           - ardana-cp1-vip-public-OPS-WEB-extapi
           - ardana-cp1-vip-public-HZN-WEB-extapi
           - ardana-cp1-vip-public-NOV-VNC-extapi
  EXTERNAL-VM:
     EXTERNAL-VM-NET: {}
  GUEST:
     GUEST-NET:
        10.1.1.2:
           - ardana-cp1-c1-m1-guest
        10.1.1.3:
           - ardana-cp1-c1-m2-guest
        10.1.1.4:
           - ardana-cp1-c1-m3-guest
        10.1.1.5:
           - ardana-cp1-comp0001-guest
  MANAGEMENT:
  ...

8.2 firewall_info.yml #

This file provides details of all the network ports that will be opened on the deployed cloud. Data is ordered by network. If you want to configure an external firewall in front of the External API network, then you would need to open the ports listed in that section.

  NETWORK NAME
     List of:
        PORT
        PROTOCOL
        LIST OF IP ADDRESSES
        LIST OF COMPONENTS

Example:

  EXTERNAL-API:
  -   addresses:
      - 10.0.1.5
      components:
      - horizon
      port: '443'
      protocol: tcp
  -   addresses:
      - 10.0.1.5
      components:
      - keystone-api
      port: '5000'
      protocol: tcp

Port 443 (tcp) is open on network EXTERNAL-API for address 10.0.1.5 because it is used by horizon

Port 5000 (tcp) is open on network EXTERNAL-API for address 10.0.1.5 because it is used by keystone API

8.3 route_info.yml #

This file provides details of routes between networks that need to be configured. Available routes are defined in the input model as part of the network-groups data; this file shows which routes will actually be used. SUSE OpenStack Cloud will reconfigure routing rules on the servers, you must configure the corresponding routes within your physical network. Routes must be configured to be symmetrical -- only the direction in which a connection is initiated is captured in this file.

Note that simple models may not require any routes, with all servers being attached to common L3 networks. The following example is taken from the tech-preview/mid-scale-kvm example.

  SOURCE-NETWORK-NAME
      TARGET-NETWORK-NAME
           default:   TRUE IF THIS IS THIS THE RESULT OF A "DEFAULT" ROUTE RULE
           used_by:
                SOURCE-SERVICE
                     TARGET-SERVICE
                     LIST OF HOSTS USING THIS ROUTE

Example:

MANAGEMENT-NET-RACK1:
    INTERNAL-API-NET:
         default: false
         used_by:
            - ardana-cp1-mtrmon-m1
            keystone-api:
            - ardana-cp1-mtrmon-m1
     MANAGEMENT-NET-RACK2:
         default: false
         used_by:
            cinder-backup:
            rabbitmq:
            - ardana-cp1-core-m1

A route is required from network MANAGEMENT-NET-RACK1 to network MANAGEMENT-NET-RACK2 so that cinder-backup can connect to rabbitmq from server ardana-cp1-core-m1

8.4 server_info.yml #

This file provides details of how servers have been allocated by the Configuration Processor. This provides the easiest way to find where a specific physical server (identified by server-id) is being used.

   SERVER-ID
         failure-zone: FAILURE ZONE THAT THE SERVER WAS ALLOCATED FROM
         hostname: HOSTNAME OF THE SERVER
         net_data: NETWORK CONFIGURATION
         state:  "allocated" | "available"

Example:

   controller1:
         failure-zone: AZ1
         hostname: ardana-cp1-c1-m1-mgmt
         net_data:
              BOND0:
                   EXTERNAL-API-NET:
                       addr: 10.0.1.2
                       tagged-vlan: true
                       vlan-id: 101
                   EXTERNAL-VM-NET:
                       addr: null
                       tagged-vlan: true
                       vlan-id: 102
                   GUEST-NET:
                       addr: 10.1.1.2
                       tagged-vlan: true
                       vlan-id: 103
                   MANAGEMENT-NET:
                       addr: 192.168.10.3
                       tagged-vlan: false
                       vlan-id: 100
         state: allocated

8.5 service_info.yml #

This file provides details of how services are distributed across the cloud.

  CONTROL-PLANE
      SERVICE
          SERVICE COMPONENT
               LIST OF HOSTS

Example:

  control-plane-1:
        neutron:
             neutron-client:
                - ardana-cp1-c1-m1-mgmt
                - ardana-cp1-c1-m2-mgmt
                - ardana-cp1-c1-m3-mgmt
             neutron-dhcp-agent:
                - ardana-cp1-c1-m1-mgmt
                - ardana-cp1-c1-m2-mgmt
                - ardana-cp1-c1-m3-mgmt
             neutron-l3-agent:
                 - ardana-cp1-comp0001-mgmt
        ...

8.6 control_plane_topology.yml #

This file provides details of the topology of the cloud from the perspective of each control plane:

control_planes:
  CONTROL-PLANE-NAME
      load-balancers:
         LOAD-BALANCER-NAME:
             address:  IP ADDRESS OF VIP
             cert-file:  NAME OF CERT FILE
             external-name: NAME TO USED FOR ENDPOINTS
             network: NAME OF THE NETWORK THIS LB IS CONNECTED TO
             network_group: NAME OF THE NETWORK GROUP THIS LB IS CONNECT TO
             provider: SERVICE COMPONENT PROVIDING THE LB
             roles:  LIST OF ROLES OF THIS LB
             services:
                SERVICE-NAME:
                    COMPONENT-NAME:
                        aliases:
                           ROLE:  NAME IN /etc/hosts
                        host-tls:  BOOLEAN, TRUE IF CONNECTION FROM LB USES TLS
                        hosts:  LIST OF HOSTS FOR THIS SERVICE
                        port:  PORT USED FOR THIS COMPONENT
                        vip-tls: BOOLEAN, TRUE IF THE VIP TERMINATES TLS
      clusters:
          CLUSTER-NAME
              failure-zones:
                 FAILURE-ZONE-NAME:
                    LIST OF HOSTS
              services:
                 SERVICE NAME:
                     components:
                        LIST OF SERVICE COMPONENTS
                     regions:
                        LIST OF REGION NAMES
      resources:
         RESOURCE-NAME:
             AS FOR CLUSTERS ABOVE

Example:

    control_planes:
    control-plane-1:
        clusters:
            cluster1:
                failure_zones:
                    AZ1:
                    - ardana-cp1-c1-m1-mgmt
                    AZ2:
                    - ardana-cp1-c1-m2-mgmt
                    AZ3:
                    - ardana-cp1-c1-m3-mgmt
                services:
                    barbican:
                        components:
                        - barbican-api
                        - barbican-worker
                        regions:
                        - region1
                                               …
        load-balancers:
            extlb:
                address: 10.0.1.5
                cert-file: my-public-entry-scale-kvm-cert
                external-name: ''
                network: EXTERNAL-API-NET
                network-group: EXTERNAL-API
                provider: ip-cluster
                roles:
                - public
                services:
                    barbican:
                        barbican-api:
                            aliases:
                                public: ardana-cp1-vip-public-KEYMGR-API-extapi
                            host-tls: true
                            hosts:
                            - ardana-cp1-c1-m1-mgmt
                            - ardana-cp1-c1-m2-mgmt
                            - ardana-cp1-c1-m3-mgmt
                            port: '9311'
                            vip-tls: true

8.7 network_topology.yml #

This file provides details of the topology of the cloud from the perspective of each network_group:

network-groups:
  NETWORK-GROUP-NAME:
      NETWORK-NAME:
          control-planes:
              CONTROL-PLANE-NAME:
                  clusters:
                     CLUSTER-NAME:
                         servers:
                            ARDANA-SERVER-NAME: ip address
                         vips:
                            IP ADDRESS: load balancer name
                  resources:
                     RESOURCE-GROUP-NAME:
                         servers:
                            ARDANA-SERVER-NAME: ip address

Example:

   network_groups:
    EXTERNAL-API:
        EXTERNAL-API-NET:
            control_planes:
                control-plane-1:
                    clusters:
                        cluster1:
                            servers:
                                ardana-cp1-c1-m1: 10.0.1.2
                                ardana-cp1-c1-m2: 10.0.1.3
                                ardana-cp1-c1-m3: 10.0.1.4
                            vips:
                                10.0.1.5: extlb
    EXTERNAL-VM:
        EXTERNAL-VM-NET:
            control_planes:
                control-plane-1:
                    clusters:
                        cluster1:
                            servers:
                                ardana-cp1-c1-m1: null
                                ardana-cp1-c1-m2: null
                                ardana-cp1-c1-m3: null
                    resources:
                        compute:
                            servers:
                                ardana-cp1-comp0001: null

8.8 region_topology.yml #

This file provides details of the topology of the cloud from the perspective of each region. In SUSE OpenStack Cloud, multiple regions are not supported. Only Region0 is valid.

regions:
  REGION-NAME:
      control-planes:
          CONTROL-PLANE-NAME:
              services:
                 SERVICE-NAME:
                     LIST OF SERVICE COMPONENTS

Example:

regions:
    region0:
        control-planes:
            control-plane-1:
                services:
                    barbican:
                    - barbican-api
                    - barbican-worker
                    ceilometer:
                    - ceilometer-common
                    - ceilometer-agent-notification
                    - ceilometer-polling
                    cinder:
                    - cinder-api
                    - cinder-volume
                    - cinder-scheduler
                    - cinder-backup

8.9 service_topology.yml #

This file provides details of the topology of the cloud from the perspective of each service:

services:
    SERVICE-NAME:
        components:
            COMPONENT-NAME:
                control-planes:
                    CONTROL-PLANE-NAME:
                        clusters:
                            CLUSTER-NAME:
                                LIST OF SERVERS
                        resources:
                            RESOURCE-GROUP-NAME:
                                LIST OF SERVERS
                        regions:
                            LIST OF REGIONS

8.10 private_data_metadata_ccp.yml #

This file provide details of the secrets that are generated by the configuration processor. The details include:

The names of each secret
Metadata about each secret. This is a list where each element contains details about each component service that uses the secret.
- The component service that uses the secret, and if applicable the service that this component "consumes" when using the secret
- The list of clusters on which the component service is deployed
- The control plane cp on which the services are deployed
A version number (the model version number)

  SECRET
      METADATA
           LIST OF METADATA
               CLUSTERS
                   LIST OF CLUSTERS
               COMPONENT
               CONSUMES
               CONTROL-PLANE
       VERSION

For example:

barbican_admin_password:
    metadata:
    -   clusters:
        - cluster1
        component: barbican-api
        cp: ccp
    version: '2.0'
keystone_swift_password:
    metadata:
    -   clusters:
        - cluster1
        component: swift-proxy
        consumes: keystone-api
        cp: ccp
    version: '2.0'
metadata_proxy_shared_secret:
    metadata:
    -   clusters:
        - cluster1
        component: nova-metadata
        cp: ccp
    -   clusters:
        - cluster1
        - compute
        component: neutron-metadata-agent
        cp: ccp
    version: '2.0'
    …

8.11 password_change.yml #

This file provides details equivalent to those in private_data_metadata_ccp.yml for passwords which have been changed from their original values, using the procedure outlined in the SUSE OpenStack Cloud documentation

8.12 explain.txt #

This file provides details of the server allocation and network configuration decisions the configuration processor has made. The sequence of information recorded is:

Any service components that are automatically added
Allocation of servers to clusters and resource groups
Resolution of the network configuration for each server
Resolution of the network configuration of each load balancer

Example:

        Add required services to control plane control-plane-1
        ======================================================
        control-plane-1: Added nova-metadata required by nova-api
        control-plane-1: Added swift-common required by swift-proxy
        control-plane-1: Added swift-rsync required by swift-account

        Allocate Servers for control plane control-plane-1
        ==================================================

        cluster: cluster1
        -----------------
          Persisted allocation for server 'controller1' (AZ1)
          Persisted allocation for server 'controller2' (AZ2)
          Searching for server with role ['CONTROLLER-ROLE'] in zones: set(['AZ3'])
          Allocated server 'controller3' (AZ3)

        resource: compute
        -----------------
          Persisted allocation for server 'compute1' (AZ1)
          Searching for server with role ['COMPUTE-ROLE'] in zones: set(['AZ1', 'AZ2', 'AZ3'])

        Resolve Networks for Servers
        ============================
        server: ardana-cp1-c1-m1
        ------------------------
          add EXTERNAL-API for component ip-cluster
          add MANAGEMENT for component ip-cluster
          add MANAGEMENT for lifecycle-manager (default)
          add MANAGEMENT for ntp-server (default)
          ...
          add MANAGEMENT for swift-rsync (default)
          add GUEST for tag neutron.networks.vxlan (neutron-openvswitch-agent)
          add EXTERNAL-VM for tag neutron.l3_agent.external_network_bridge (neutron-vpn-agent)
          Using persisted address 10.0.1.2 for server ardana-cp1-c1-m1 on network EXTERNAL-API-NET
          Using address 192.168.10.3 for server ardana-cp1-c1-m1 on network MANAGEMENT-NET
          Using persisted address 10.1.1.2 for server ardana-cp1-c1-m1 on network GUEST-NET

        …
        Define load balancers
        =====================

        Load balancer: extlb
        --------------------
          Using persisted address 10.0.1.5 for vip extlb ardana-cp1-vip-extlb-extapi on network EXTERNAL-API-NET
          Add nova-api for roles ['public'] due to 'default'
          Add glance-api for roles ['public'] due to 'default'
          ...

        Map load balancers to providers
        ===============================

        Network EXTERNAL-API-NET
        ------------------------
          10.0.1.5: ip-cluster nova-api roles: ['public'] vip-port: 8774 host-port: 8774
          10.0.1.5: ip-cluster glance-api roles: ['public'] vip-port: 9292 host-port: 9292
          10.0.1.5: ip-cluster keystone-api roles: ['public'] vip-port: 5000 host-port: 5000
          10.0.1.5: ip-cluster swift-proxy roles: ['public'] vip-port: 8080 host-port: 8080
          10.0.1.5: ip-cluster monasca-api roles: ['public'] vip-port: 8070 host-port: 8070
          10.0.1.5: ip-cluster heat-api-cfn roles: ['public'] vip-port: 8000 host-port: 8000
          10.0.1.5: ip-cluster ops-console-web roles: ['public'] vip-port: 9095 host-port: 9095
          10.0.1.5: ip-cluster heat-api roles: ['public'] vip-port: 8004 host-port: 8004
          10.0.1.5: ip-cluster nova-novncproxy roles: ['public'] vip-port: 6080 host-port: 6080
          10.0.1.5: ip-cluster neutron-server roles: ['public'] vip-port: 9696 host-port: 9696
          10.0.1.5: ip-cluster heat-api-cloudwatch roles: ['public'] vip-port: 8003 host-port: 8003
          10.0.1.5: ip-cluster horizon roles: ['public'] vip-port: 443 host-port: 80
          10.0.1.5: ip-cluster cinder-api roles: ['public'] vip-port: 8776 host-port: 8776

8.13 CloudDiagram.txt #

This file provides a pictorial representation of the cloud. Although this file is still produced, it is superseded by the HTML output described in the following section.

8.14 HTML Representation #

An HTML representation of the cloud can be found in ~/openstack/my_cloud/html after the first Configuration Processor run. This directory is also rebuilt each time the Configuration Processor is run. These files combine the data in the input model with allocation decisions made by the Configuration processor to allow the configured cloud to be viewed from a number of different perspectives.

Most of the entries on the HTML pages provide either links to other parts of the HTML output or additional details via hover text.